Skip to main content

Full text of "Electron microscope"

See other formats


S " 

X cJ 


Shown below is a scale of the dimensions of the universe as 
atom's nucleus, as the universe's inside dimension, to the dis- 






















I ! II 






H K 





so far measured by man. It ranges from the diameter of an 
tance to the farthest galaxy recorded by astronomers. 































Courtesy Life Magazine 




Courtesy University of Toronto 

Wing of Midge (Chironomidae) , X 19,400. Electron microscope picture taken through 
the wing of a small insect called a midge or gnat. It shows structures on the wing which 
could never be revealed by any optical microscope. The "hairs" on the wing are distributed 
at the rate of over 400 million per square inch. What their function is no one at present knows. 







Head of the Department of Physics 

and Director of the McLennan Laboratory 

University of Toronto 



Formerly Chief Engineer 

Rogers Electronic Tubes Limited 


Drawings by Dorothy Stone 





c ,2- 

Copyright, 1946, by 

All rights reserved 

Printed in the United States of America by 


Preface to First Edition 

"I was received very kindly by the warden [of the Grand 
Academy of Lagodo], and went for many days to the academy. 
Every room hath in it one or more projectors [researchers] .... 

"The first man I saw was of a meager aspect with sooty hands 
and face, his hair and beard long, ragged and singed in several 
places. His clothes, shirt, and skin were all of the same color. 
He had been eight years upon a project for extracting sunbeams 
out of cucumbers, which were to be put into vials hermetically 
sealed, and let out to warm the air in raw, inclement summers. He 
told me he did not doubt, in eight years more, he should be able to 
supply the governor's gardens with sunshine at a reasonable rate; 
but he complained that his stock was low, and entreated me to 
give him something as an encouragement to ingenuity, especially 
since this had been a very dear season for cucumbers. I made 

him a small present " Gulliver's Travels, Voyage 

to Laputa, etc., Chapter V. 

With the exception that sartorial and sanitary arrangements 
may have advanced with our civilization, there is an uncanny par- 
allel between the "projector" of Lagodo in 1741 and the electron 
microscope investigators at Toronto in 1941. We have u been 
eight years upon a project"'' for taking magnified pictures without 
light, an investigation which is, to the modern Gulliver, no less 
foolish than the cucumber experiment was to the original Gulliver. 

We "do not doubt, in eight years more," that there will be an 
electron microscope in every "project" laboratory and that it will 
contribute greatly to our knowledge of things as remote from 
each other as industry and medicine. 

A third item in the parallelism is apparent, namely "that our 
stock is low," and we still entreat contributions "as an encourage- 
ment to ingenuity," especially since our "cucumbers" are still 
expensive. In spite of the fact that the electron microscope bids 


fair to reveal to the eyes of man in the domain of the very minute 
things quite as wonderful as Galileo's telescope revealed of the 
wonders of space, we still have the feeling that it is ever "a dear 
season for cucumbers." 

E. F. Burton 
W. H. Kohl 
Toronto, Ontario 
March 1, 1942. 


(from first edition) 

First in the list of acknowledgements, I wish to thank my co- 
author, Dr. W. H. Kohl, for the great help he gave the Depart- 
ment of Physics of the University of Toronto, at the beginning 
of the work on the electron microscope. It is on account of this 
assistance that Dr. Kohl was the natural person to call upon for 
co-operation in the preparation of the present book. For many 
years he has been a member of the staff of the Research Labor- 
atory of the Rogers Radio Tubes Limited at Toronto, and for 
the greater part of this time he has served as special lecturer at 
the University. 

His first course of lectures was on electron optics and, as he 
was engaged in development work on cathode ray tubes in an 
industrial laboratory, he was in a position to repeat some of 
Briiche and Johannson's experiments on the electrostatic electron 
microscope and produce images of oxide cathodes which revealed 
helpful information on the structure of the coating. Such an 
electron microscope was demonstrated before the seminar of the 
McLennan Laboratory in April, 1934; this probably was the first 
such demonstration in Canada. Images of oxide cathodes pro- 
duced with magnetic lenses were demonstrated during a special 
course of lectures on electron optics in January, 1935. 

Active research work was begun in this field at the Toronto 
laboratory during the fall term of 1935, by C. E. Hall, who con- 
structed a simple electrostatic electron microscope and repeated 
quantitative measurements of Johannson on the properties of the 
immersion objective. In 1937 Mr. Hall joined the research lab- 
oratory of the Eastman Kodak Company, Rochester, New York, 
where he continued work in this field. 

In the fall of 1937 J. Hillier, B.A. (Toronto), and A. Prebus, 
B.A. (Alberta), undertook the construction of a high-voltage 
magnetic compound microscope with the aim of applying it to the 



investigation of biological specimens. Their work was highly 
successful and was first described in the Canadian Journal of 
Research in April 1939. 

In February 1940, Mr. Hillier joined the laboratories of the 
R.C.A. Manufacturing Company and proceeded to design, in 
collaboration with Mr. A. W. Vance and under the direction of 
Dr. V. K. Zworykin, a type of electron microscope adapted to 
industrial research. Mr. A. Prebus joined the Ohio State Univer- 
sity in July 1940. To Messrs. Hall, Hillier and Prebus we are 
greatly indebted, and also to Professor Arnold Pitt. 

The work in Toronto is now being carried on by Mr. W. E. 
Ladd, Dr. L. T. Newman and Mr. J. H. L. Watson, who have 
assisted in the construction of a second instrument. 

For financial assistance and encouragement, I wish to thank 
very heartily the following: 

The National Research Council of Canada for direct grants in 
aid and scholarships. 

The Banting Department of Medical Research of the Uni- 
versity of Toronto, through the interest of the late Sir Frederick 
Banting and his staff. 

The Banting Foundation for Medical Research for grant in 
aid toward work on viruses being carried out with the excellent 
co-operation of Dr. J. Craigie of the School of Hygiene, Univer- 
sity of Toronto. 

The Columbian Carbon Company of New York City and par- 
ticularly Mr. W. B. Wiegand, their Director of Research, and 
the members of his staff, Mr. H. A. Braendle and Dr. Carl Sweit- 
zer; this financial aid came at a time when it was needed most. 
It is pleasant to recall that all three are former students of the 

The Ontario Mining Association for a grant for the study of 
mine dust as related to silicosis. This work was undertaken 
through the interest of Dr. D. A. Irwin of the Banting-Best Medi- 
cal Research Department. 

The President and Board of Governors of the University for 
sympathetic assistance at all times. 


The staff of our workshop and glass blowing plant, through 
whose ability and energy the successful construction was made 
possible. Two complete instruments were entirely produced in 

our own shops. 

E. F. Burton 

Preface to Second Edition 

This book addresses itself essentially to the non-technical 
reader who is nevertheless interested in modern techniques. It 
aims at presenting the physical principles upon which the opera- 
tion of the electron microscope is based, without any assumptions 
in regard to the technical knowledge of the reader. The subject 
matter is such, however, that a large amount of ground must be 
covered and many intricate problems solved. This requires more 
than a superficial interest and a great deal of patience and per- 
severance on the part of the reader. In return, much useful 
knowledge may be gained, which applies to general physics and 
electronics. Thus it is hoped that the book will appeal not only 
to laymen and advanced high-school students, but also to mem- 
bers of various professions, in particular those in which the elec- 
tron microscope has made, or promises to make, a considerable 
contribution in specialized research. 

The second edition has been prepared at the suggestion of the 
publishers. The text has been re-arranged and revised in many 
places to attain greater coherence in the presentation of the sub- 
ject matter. In this the authors are obliged to critical reviewers 
of the first edition for helpful criticism. The changes are too 
many to be enumerated here but they will be apparent from the 
table of contents, if this is compared with that of the first edition. 

In order to indicate the aim of the book a subtitle has been 
added to the original title. Some material has been eliminated as 
being of too specialized a character for the scope of this book. 
This applied in particular to the detailed description of the 
Johannson-Briiche electrostatic electron microscope. The chap- 
ter on the history of the electron microscope has been eliminated 
and part of its content incorporated in other chapters. The 
chapters on electron optics have been enlarged and the material 
on electron lenses consolidated therein. The chapter on the 
compound magnetic electron microscope has been re-written 



completely and much new material added. The description of 
iron-clad magnetic lenses was removed from this chapter and 
included in the chapter on magnetic lenses, where it properly 
belongs. The chapter on the applications of the electron micro- 
scope has also been largely rewritten and the plate illustrations 
of the accomplishments of the electron microscope are collected 
here. Some plates were eliminated and new ones added. A com- 
plete bibliography, prepared and published recently by Claire 
Marton of Stanford University and Samuel Sass of the University 
of Michigan, is included with the kind permission of the authors 
and the publishers of the Journal of Applied Physics; this has 
been added to by references to the more recent papers. In order 
to emphasize the importance to which the electron microscope 
has grown on this continent, some data have been included on 
the organization of the Electron Microscope Society of America. 

While both authors have been handicapped in their efforts by 
the unusual load of pressing war work, it is hoped that this second 
edition will meet with some favor among those interested in this 
young branch of physics, which promises to grow ever stronger 
and yield fruit beyond the fondest expectations of pioneers in 
this field. 

The authors wish to thank those who have taken part in the 
work on the electron microscope at Toronto during recent years: 
namely, Mrs. B. M. Deacon, Ph.D., Dr. L. T. Newman, Dr. J. H. 
L. Watson and Mr. S. G. Ellis, who helped considerably in the 
preparation of the later chapters. The 1944 Toronto microscope 
was constructed in our own workshop, almost entirely by Mr. 
Grantley Woodward. 

We also wish to acknowledge with gratitude the continued 
assistance given by the National Research Council of Canada 
and also support from the Shawinigan Chemicals Limited, Shawin- 
igan Falls, Quebec and the Imperial Oil Company, Limited, of 
Sarnia, Ontario. 

E. F. Burton 
W. H. Kohl 

Toronto, Ontario 

December, 1944 



Preface to First Edition 3 

Acknowledgements from First Edition 5 

Preface to Second Edition 9 


1. Vision 13 

2. Light Microscopes 25 

3. What Is Light? 43 

4. Wave Motion and Wave Motion Media 53 

5. Wave Motion in Elastic Media 59 

6. Light Waves and the Microscope 67 

7. The Electromagnetic Theory of Light 81 

8. The Electron 87 

9. Electron Emission 99 

10. The Dual Theory of Light 109 

11. The Dual Theory of the Electron 121 

12. The Motion of Electrons in Electrical Fields 127 

13. The Motion of Electrons in Magnetic Fields 139 

14. Electron Optics I. Magnetic Focussing 159 

15. Electron Optics II. Electrostatic Focussing 175 

16. The Electrostatic Electron Microscope 191 

17. The Magnetic Electron Microscope 207 

18. The 1944 Toronto Microscope 221 

19. Useful Magnification 231 

20. The Use of the Electron Microscope 239 

21. Practical Applications of the Electron Microscope 251 

Epilogue 298 

Bibliography of Electron Microscopy 299 

Addenda to Bibliography 318 

Indexes 319 


Courtesy University of Michit 

Portland Cement (hydrated 18 days), X 20,000. 

Chapter 1 

The human eye is a wonderful optical instrument, more 
wonderful than any that man has ever devised. It unfolds to 
our mind the world that surrounds us and is the link which 
enables us to take a conscious part in it. But fascinating and 
manifold as is the scene which the naked eye presents, it does 
not satisfy our curiosity. We like to see things which the unaided 
eye cannot see, to bring far distant and ordinarily invisible stars 
within our view and to penetrate into the realm of the very small 
things, the microcosmos. For this purpose we need artificial aids: 
telescopes in the field of astronomy, and microscopes in the study 
of the very minute. 

Whenever we interpose an artificial optical system between 
the eye and an object we are attempting merely to assist our eyes 
so as to widen their scope of vision. The eye is always the indis- 
pensable part of a sequence: object — vision aid — eye. In this 
book we are particularly concerned with the microscope. We 
must then keep in mind that it is only a link in this chain which 
we have just described. 

Two almost self-evident conditions of vision must be recalled 
at the very beginning; first, an object is made visible by the light 
which it diffuses, scatters or reflects, and secondly, the position 
of the object viewed is judged to be in the direction of the line 
along which the light finally enters the eye. This latter circum- 
stance lies at the basis of most optical illusions. 

The Bending of Light Rays: Refraction 

The following simple parlor trick illustrates quite well what 
happens to a beam of light as it passes from one transparent 
medium to another. Place a coin on the bottom of a mug and 
see that the coin is well enough illuminated to be seen when 




viewed from above. Have an observer keep his head in such a 
position that he just can't see the coin as he peers over the edge 
of the mug (Fig. la). If, while the observer keeps his eye fixed 

The angle A greater 
than angle 73' 



Mug filled \==-ZT%==\ 
ujifh ujater 

Fig. la. Fig. lb. 

How light bends as it passes from water to air. 

in position, the mug is filled with water, the coin comes into view 
(Fig. lb). The broken straight line represents the path along 
which the light appears to travel from the coin to the eye. But 
we know that the coin has not changed its position. Therefore 
the real path of the light from the coin to the eye must be that 
marked by the dotted line. 

'•I'-v.'/v* ':•■':' 

:•■■:....•'■ :„•. .; 

Apparen t position 
of sun 

Real position 
of sun 

Fig. 2. How one can see the sun after it has set. 

Another example of such an optical illusion, on a grand scale, 
is the fact that we see the sun at sunset for some considerable 



time after it has really sunk below the horizon. The reason for 
this is that the light from the sun is gradually bent around in going 
through layers of the atmosphere of differing densities as shown in 
Fig. 2. Our judgment of the position of the sun is dictated by 
the direction along which the light finally enters the eye. 

These two phenomena are perhaps the earliest recorded obser- 
vations relating to light; they were described by Cleomedes in 
A.D. 50. 

The Direction of Bending of a Light Beam 

Very cursory examination of Fig. lb tells us how the light 
bends as it passes from water to air: the angle which the emergent 
ray makes with the side of the mug, or rather with the perpen- 
dicular to the surface of the water, is larger than the angle which 
the ray of light from the coin makes with that perpendicular. 
In general, we may say that when light is going obliquely from a 
dense medium (water) into a less dense medium (air) it bends 
away farther from the normal to the surface of separation of the 

Fig. 3. 

How light bends as it passes 
from air into water. 1% J 

The angle '3 'is less than 
angle A' 

We can also perform the reverse experiment. If we direct a 
narrow beam of light, say from a flash-light, obliquely onto the 
surface of some slightly turbid water contained in a glass bowl, 


we can see easily that, as the light goes from the air (the less dense 
medium) into the water (the denser medium), the direction of 
the ray is bent nearer to the normal to the surface (Fig. 3). 
This change of direction which a light beam undergoes when it 
passes from one transparent medium into another is described 
by the term refraction. 

The Prism and the Lens 

We are all familiar with effects produced by blocks of glass 
or automobile lamp lenses in causing a change in the direction of 
light passing through them. Fundamentally all such effects can 
be simply illustrated by tracing a ray of light through a glass 
prism, which is a uniform block of glass with a triangular cross- 
section (Fig. 4). 

Consider a candle placed at the point, P, and one's eye at the 
point, E. If we interpose a glass prism, ABC, in our line of 
vision, EP, we would judge that the candle is displaced to the 
position Q. Light travelling along the line PE will now be turned 
off the track at K and will not reach the eye. The ray which 

B C 

Fig. 4. How a light beam is deviated by a prism. 

does reach the eye is one travelling from the candle in some such 
direction as PG. At the point G, as the light is passing from a 
rare medium (air) to a dense medium (glass) its direction of 
travel inside the prism makes an angle with the normal to the 


surface at the point G, which will be less than the angle XGP; 
and then as it passes out of the glass again into the air the angle 
YHE will be larger than that which the ray, GH, makes with 
the normal to AC. The eye then judges the candle to be along 
the direction EHQ, which is the direction of the straight line 
along which the light enters the eye. The prism has thus caused 
a distinct change in the direction of the ray which ultimately 
enters the eye. 

After this account of the deviation of a beam of light caused 
by a prism, it is easy to understand the action of a simple lens, 
such as a reading lens, when it is used to "focus" on a screen the 
image of a distant source of light or a brightly illuminated object. 

It is quite common practice to interpose such a lens between 
a bright object, (O, Fig. 5), and a screen, e.g., one's hand, and to 
produce on such a screen an image as indicated at I. The position 
of the lens and the screen always has to be adjusted in order to 
obtain the position of I where the image is sharp. If the experi- 
ment is performed in a darkened room and the source is shielded 

° *"^£::;;' 

Fig. 5. How a lens bends light. 

so as to send out light only in the direction of the lens, the actual 
path of the light from O to I can be seen, particularly if some 
smoke or chalk dust is blown into the region round the lens. 

The reproduction at I is called the image of the object O, and 
the process of obtaining a sharp image by adjustment of the 


position of the lens and screen is spoken of as focussing the lens. 
The explanation of this action of the lens is quite clear when 
we realize that any lens can be considered as. being built-up of a 
great number of portions of prisms, as indicated in Fig. 6. By 
following the paths of rays from O through the various prism 
sections, it is apparent that it is quite possible for the emergent 
rays all to pass through the point I. 


Fig. 6. How a lens may be considered as made up of sections of prisms. 

The greater the number of prism sections that we put in Fig. 6, 
the more nearly the outline of this prism structure approximates 
that of the ordinary lens. The surfaces of lenses are usually 
spherical. We can, however, draw the path of a ray through 
such a lens without recourse to the prism structure. 

Fig. 7 illustrates the path of a ray of light from air through 
a single surface of glass which has spherical curvature. The 
construction is made just as in Fig. 4; the normal to the glass 
surface, N, is along the direction of the radius of the curve. 

M ^ r r' /'//M////. 

n . ^.M^Vi/ Refracted Ray FlG - 7 - 

Hlf ^.»* \'// / yy(/ / / / P at h °f rav through a single sur- 

# »*f \// ' / / C / / / ^ ace °f gl ass > w i tn spherical curvature. 

Similar construction is carried through for the two curved sur- 
faces of a lens in Fig. 8. The amount of deviation of the incident 


ray, OPi, caused by a curved surface depends on the curvature of 
the surface, or, in other words, on the length of the radius of 
the curve; for example, on the lengths of C1P1, and C 2 P 2 in 
Figure 8. The shorter the radius, the greater the curvature of the 
surface, and the greater the deviation of the ray. This deviation 
fixes what is known as the power of the surface or of the lens. 
Through the action of such lenses we can explain the function- 
ing of the eye and the role played by the microscope as an aid 
to vision. 

N- D 

<** [Ai</A o\ 




Air \ '/ 6 A // V 


Fig. 8. Path of ray through two spherical surfaces of a lens. 

The Eye 

The eye is nature's camera, the prototype of all cameras. 
The structure and function of the various parts of the eye can 
be very well understood by referring to the corresponding parts 
of an ordinary camera. 

The simplest possible form of camera is the so-called pin-hole 
camera — a light-tight box with a pin hole in the center of one side 
and a photographic plate or film placed against the inner side 
directly opposite the hole (Fig. 9). 

If the box is turned with the pin hole toward an object, such 
as the animal at AD, and the hole is uncovered so as to let light 
into the box, an image of the animal will be impressed on the plate 
or film. This image will be produced in the following manner. 
Each part of the animal that faces the camera sends light into the 
pin hole, the aperture. The amount of light which is diffused by 
any portion of the surface of the animal depends on the natural 
color of the surface. Practically no light will come from black 



patches and a great deal of light from white patches. Since light 
travels in straight lines from any surface point on the animal to 
the aperture, all rays will cross one another in the aperture and 
continue on to the photographic plate. Thus the light ray com- 
ing from a point A on the head of the animal will impinge on the 
film at point A' and blacken the film according to its intensity. 
The distribution of light and shade at the surface of the object 
will thus be accurately reproduced on the film and a picture of 
the animal obtained on developing the film. 



Fig. 9. A pin-hole camera. 

The pin-hole camera, though interesting as a hobby and 
instructive as a scientific experiment, has its limitations. The 
amount of light entering the pin hole is so small that a long expo- 
sure is always necessary and so the use of this camera is limited to 
still objects. As the distance from the pin hole to the film is 
fixed there is no means of adjusting for variable focus, and conse- 
quently no one part of the picture is more sharply in focus than 
any other. 

We now turn to the simple camera, which is familiar to all 
(Fig. 10). It consists of a light-tight box provided with a con- 
vergent lens through which the light enters, and a plate or film 
on which the image formed by the lens can be focussed. Since 



the light-tight box is furnished with a bellows body the lens-film 
distance may be adjusted without letting in any extraneous light. 
The advantage of such a camera over the pin-hole type is that 
more light from any point on the object is concentrated at the 
corresponding point on the image, and therefore a much smaller 
exposure is necessary in order to cause a definite effect on the 
film. It is consequently possible to take "instantaneous" exposures 
of moving objects. 

A '• 






Fig. 10. An ordinary camera; at the right is a representation of what is seen in 
looking into the back of the box. 

The functioning of the human eye may now be readily under- 
stood, at least in its basic principles, from what we have said about 
the working of a camera. 

The eye-ball is a light-tight spherical box with a great part of 
the inner surface covered with a sensitive film, called the retina, 
from which light impressions are transferred to the brain (Fig. 
11). Just as in the case of the camera, the relation between the 
positions of the lens, the object and the image must be adjustable 
in order to obtain a sharp image on a light-sensitive surface, the 
retina. With the eye, however, these adjustments are not made 
by changing the position of the lens but by the individual adjust- 
ing the size and shape of the lens itself, so that the image is always 
formed on the retina, which remains at a fixed distance from the 
pupil of the eye. If we wish to look at a distant object we auto- 
matically adjust the eye lens so that the image of the distant 



object will be focussed on the retina; as the object comes nearer 
and nearer the eye lens is adjusted through muscular control 
until we see the object distinctly in any new position. A normal 
person sees with the greatest comfort when the eye is allowed to 
rove over distant objects and no effort is exerted to bring any 
particular thing into focus. When one concentrates on carefully 
viewing near objects, the eye tends to become tired and suffers 
from strain. It is impossible for the normal eye to focus on 

Optic nerve 

Fig. 11. How the eye forms an image. 

objects nearer than about 10 inches (25 cms.). This distance is 
known as the least distance of distinct vision and represents a 
very important natural limit of vision. No matter how we try 
to help the eye by artificial aids, say by a microscope, we must 
end up with at least the illusion that we are observing the objects 
or images as though they were placed 10 inches from the eye. 

The Scope and Limitation of the Eye 

The normal eye has a very wide field of vision and a great 
range in distance. It will be apparent from Fig. 12 that, as the 
object recedes from the eye, the area of the retina covered by the 
image, e.g., at A'B', becomes smaller and smaller, and finally, as 
the distance increases very greatly, we say the object disappears. 
Light does not cease to come from the object, but the area of the 
retina affected by the light coming from the object becomes so 
small that, for some physiological reason, it ceases to produce a 
sensation of vision in the brain. It has been found that, if the 



angle subtended at the eye by the object becomes less than about 
1.4 minutes (about two one-hundredths of a degree), the object 
ceases to be distinguishable. 

5* \^i 

PUt 0f~S?SAt _ 

Fig. 12. How the size of the image on the retina decreases as the object is moved 
farther and farther from the eye. 

From this it follows that whether any particular object is 
visible to the eye or not depends on both its size and its distance 
from the eye; in any case the visibility is limited by the angle sub- 
tended at the eye (see Table 1). Of course, we take for granted 
that the object is always brightly enough illuminated to make 
vision possible within the limits just expressed. 
Table 1. Sizes of Objects Just Visible to the Eye at Different Distances. 

Distance of Object 
from the Eye 

10 miles 
1 mile 
100 yards Lin ^ r s ^J}^ ion 
10 feet Object Visible 
10 inches 

20 feet 

2 feet 

1 inch 

1/20 inch 

1/250 inch, 

4 mils, 

1/100 cm., or 1/10 mm. 

Now an object may be so small that it is invisible at a distance 
of 10 inches from the eye; this is true if it is so small that it sub- 
tends an angle at the eye of less than 1.4 minutes. Since the eye 
cannot focus on anything which is closer to it than 10 inches, we 
must conclude that the unaided ("naked") eye cannot see any 
object less than l/250th of an inch in diameter, nor can it make 
out any detail of any marking on a large visible object if the 
linear dimension of the pattern is less than l/250th of an inch. 

In this extremity we call on the microscope to assist the eye. 









(Above) Diatom, X 36,000. The original electron micrograph was taken at a magnification 
of X 12,000 and afterwards enlarged in the ordinary optical manner by X3. Courtesy Ameri- 
can Cyanamid Co. 

(Below) View of image of grease specimen on fluorescent screen used for final focussing, 
made with Toronto 1944 microscope. Courtesy University of Toronto. 

Chapter 2 

Aids to Defective Vision 

Everyone is fairly familiar with the simplest cases of defective 
vision, such as short-sightedness and long-sightedness, and the 
means used to correct such defects. In one case, the eye is able 
to focus the image distinctly on the retina only when the object is 
held abnormally near the eye; in the other case, the object must 
be held abnormally far from the eye. If the position of the object 
is at the normal distance for distinct vision, i.e., ten inches, then 
for such defective eyes the image is formed either in front of the 
retina or behind it. In either case the object is not seen distinctly 
(see Fig. 13). 

These defects are corrected by the use of eye-glasses or 
spectacles; for the case illustrated in Fig. 13a the eye-glass throws 

Fig. 13a. Near-sightedness. 

Fig. 13b. Far-sightedness. 

the image farther away from the lens of the eye on to the retina 
(Fig. 14a), while, for that in Fig. 13b, the convergence of the 



light is increased as it enters the lens of the eye and thus the 
image is brought back to the retina (Fig. 14b). 


— -:.•*, 

Fig. 14. 

How glasses correct near- 
end far-sightedness. 

Sfyv; « 


How to Tell the Position of the Image Formed by a Single Lens 

Before explaining further the operation of these aids to vision, 
we shall have to present a few simple rules which enable us to 
estimate by means of a diagram the position of images formed by 
any lens. Simple lenses are of two general classes, converging 
and diverging, depending on what happens to a beam of light 
falling on the lens. This can be illustrated experimentally by 
observing what happens to a beam of sunlight, say, admitted 
through a hole in a window blind, when the beam falls on the lens. 
A simple application of the principles described in the first 
chapter regarding the bending of the light ray as it goes from 
dense to rare or from rare to dense media will justify the repre- 
sentation given in Figs. 15 and 16. 

In the case of the converging lens the rays parallel to the 
axial line XY all pass through a point F on the axis; this point 

Fig -15. x ^~^^[ ^^{i^^^ Y 

Converging Lens. *".",]'.', *[ .['.*"*.'," ' 

is called the focus of the lens. If the beam of parallel rays were 
sent through the lens in the opposite direction, from Y to X, 
there would be found a similar point, F', on the opposite side of 
the lens; this is also called a focus. If the lens is symmetrically 


shaped on the two sides, the distance of F from the center of the 
lens will be the same as the distance of F' from the center; this 
distance is called the focal length of the lens. 

Rule 1: Every ray entering the lens parallel to the axis XY 
passes out through the focus F, or F, as the case may be. 

Using a beam of sunlight, if a screen or the palm of the hand is 
held at the focus F, or F', a sharp point of light will be seen — in 
reality an image of the hole in the window blind. 

If a diverging lens is used no such image will be formed on a 
screen placed anywhere; but, if the eye is placed in the position 
represented in Figure 16, the observer will have the illusion that 

Fig. 16. 

Diverging Lens. 

the light seems to come from a point F' on the side of the lens 
remote from the eye. This point F' is called the focus of the 
diverging lens; so for this case Rule 1 becomes: Every ray enter- 
ing the lens parallel to the axis XY passes out in a direction as 
though it proceeded from the point F, the focus of the lens. As 
in the case of the converging lens, if the diverging lens is sym- 
metrical as to its two sides there will also be a similar focal point 
on the other side of the lens at the same distance from the center 
of the lens as F'. 

There is a second rule applicable to both kinds of lenses. 
Rule 2: Every ray entering the lens at any inclination and pass- 
ing through the center point of the lens 'will emerge with no per- 
ceptible deviation from its original direction (Figs. 17a, 17b). 
This is strictly true only for very thin lenses, but it is permissible 
to use it when one is just approximating to the real conditions. 

The two rules given above enable us to find the position of the 
image of any object very easily (Fig. 18). The image of A will 
be on the line AO produced through the lens and also along PF 
produced; these lines meet in the point A' which is consequently 
the image of the point A. Similar procedure gives B' as the image 






Fig. 17a. Fig. 17b. 

For a thin lens the ray through the central point is not deviated perceptibly. 

of B; consequently the figure A'B' marks the image of AB, since 
all points on the object between A and B will produce image 
points in corresponding positions between A' and B'. 

It is important to note that the image A'B' produced by the 
biconvex lens (Fig. 18) is inverted, both vertically and hori- 
zontally, in relation to the position of the object, AB, and that 
the image is larger than the object. In other words, the object 
has been enlarged or magnified by the lens. This principle is 
applied in the lantern slide projector where the enlarged image 
is thrown upon a screen. The slide must be inserted in the slide 
carrier upside down and left to right in order that the image may 
be presented in the proper aspect. This rule is only too frequently 
neglected; many a potentially first-class lecture has been spoiled 
by inexperienced lantern operators. 


Fig. 18. How to determine graphically the position of the image of an object when 
using a single converging lens. 


The Simple Microscope, Reading Lens or Magnifying Glass 

It has been pointed out above that the unaided eye cannot 
see objects distinctly when they are at a distance less than 10 
inches from the eye and that it ceases to perceive small objects 
even at the optimum distance of 10 inches when these objects 
are less than l/250th of an inch in linear dimension. 

It is in the latter circumstance that we call to our aid the 
magnifying power of lenses such as are illustrated in Fig. 18. 
The eye itself contains a single converging lens. Fig. 19 shows 
what happens to the image of an object as the object is brought 
closer and closer to such a lens. 

• •••••••• •••••< 



Fig. 19. How the position of the image changes as the object is moved closer and 
closer to a converging lens. 

The ordinary rules give the positions of the images of AA 
and A 2 B 2 at A'^ and A' 2 B' 2 respectively. However, when we 
follow the two standard rays from the point A 3 of the object 
A 3 B 3 the light emerges from the lens along the lines PF and A s O, 
which never meet to the right of the lens; but if the eye is placed 
to the right of the lens and close to it (Fig. 20) the observer will 
have the impression that the light appears to come from the 
position A' 3 B' 3 . Thus to the eye the object is enlarged and the 
position of the image must be at least ten inches from the eye in 
order to be in focus. This shows why we have to adjust the 
relative positions of the eye, the lens and the object in using such 




Fig. 20. How the eye sees the i 


a simple microscope. It is at once apparent that the object must 
be placed between the focus of the lens and the lens itself. 

The Compound Microscope 

The ordinary form of a compound microscope merely uses 
two individual lens systems to produce magnification in two 

There are really two fundamental forms of the compound 
microscope; the first is represented by a combination of two sep- 
arate systems, each similar to that shown in Fig. 18, and the sec- 
ond by a combination of two separate systems, as shown by com- 
bining Figs. 18 and 20. 

For example, we may produce an enlarged image, A'B', of an 
object, AB, by a simple lens (or single lens system) as in Fig. 18 
and then, using this image as a new object, produce a still larger 
image by a second lens, as illustrated in Fig. 21. This new image, 
A"B ', can be thrown on a screen or frosted glass and examined 
by the eye; it can also be projected on a photographic film or 
plate; thus we may obtain a permanent record of the enlargement. 
The first lens or system, L u is known as the objective and the 
second lens system, L 2 , as the projector. It is apparent that the 
second image, A"B", can be treated again as a new object, and a 
still larger image may thus be obtained by another projector lens 

The common usage of the term compound microscope refers 
to an instrument used directly by the eye. One looks into the 



Eye looking 
at screen 


t^*'"'. IV': B .'1 / •• 

L Objective 

Photographic plate, 
or frosted screen 

Projecting Lens 

Fig. 21. A compound microscope as used to project or photograph the image. 

eye-piece at the top of the instrument and sees the image. Fig. 22 
shows diagrammatically the lay-out of the component lenses. 
The objective, L 1; forms the first image, A'B', as shown in Fig. 21. 
This image, A'B', now serves as the object for the eye-piece which 
acts merely as a simple microscope. Consequently, the right-hand 
portion of Fig. 22 is merely a repetition of Fig. 20, and, as 
described under the section on the simple microscope, the eye gets 
the impression that the light is proceeding from the image, A"B", 
at a distance of ten inches from the eye. 



lye piece 


• *"T " **• • • • 
•••' A" 

;':';V t 

10 inches 

Fig. 22. An ordinary compound microscope for visual observation. 


The lens of the eye gathers in the slightly diverging beam 
which leaves the eye-piece and focusses this beam on the retina. 
Of course, this same eye-piece might have been used on the image, 
A"B", of Fig. 21, and the eye would have seen a final image of 
much greater magnification. & 

In fact theoretically there is no limit to the magnification that 
one might attain by using lens after lens after lens. However, in 
practice, there are limitations in construction and lighting which 
make a progressive magnification, ad infinitum, impossible. Some 
of these difficulties are noted in the next section. 

Aberrations of Lenses and Lens Systems 

In the preceding discussion we have tacitly assumed that all 
the lenses acted in an ideal fashion, that is, that the lens produced 
in the image a true representation of the object, point for point. 
But there are many ways in which actual practice falls far short 
of this ideal. These shortcomings are known by the general name 

If we look back at Figs. 5 and 18 we will find that we have 
assumed that light emanating from an object point is sharply 
focussed at a definite corresponding point in the image. But 
this is not so if we use only one lens. For one thing, the outer 
annular zone of the lens does not cause the light from a point 
to converge to exactly the same point as does the central zone; 
this defect of the lens is known as spherical aberration. Again,' 
if we use ordinary white light, which as we know is just a mixture 
of light of different colors, we find that the different colors do 
not all converge to the same point. As a result the images are 
fringed with color— a defect known as chromatic aberration. 

These are only two of the shortcomings of lenses. In order to 
overcome such defects, it is necessary to use a system of several 
individual lenses of differing shapes and made of different kinds 
of glass, instead of one single lens, as indicated in our figures. It 
would be quite beyond the scope of this book to deal at all fully 
with the subject of aberrations and their corrections. In fact, 
no matter how elaborate the system of lenses which makes up the 


objective or the eye-piece of an ordinary compound microscope, 
we may treat each system as equivalent to a single converging 
lens as far as our present needs require. But we must remember 
that striving for perfection in the reproduction of an object in- 
troduces complexity into the optical systems in theory as well as 
in practice; attempts to produce a theoretically perfect micro- 
scope add greatly to the cost of a fine instrument. 

The Resolving Power of a Microscope 

The purpose of any microscope may be looked upon as either 
to enable us to see objects so small that they cannot be dis- 
tinguished by the naked eye or to show us the finest detail m a 
large object. These two features really amount to the same thing, 
because if a small detail, (Fig. 23), is visible in a microscope 
it is because two points, which are separated by a distance equal 
to A-B on the object, can be distinguished as separate points. We 
have seen that for the unaided eye this limiting distance is about 
l/250th of an inch. The limiting distance for any given micro- 
scope is known as its resolving power. 

Fig. 23. 

^ g Separation of points and 

• • linear dimensions of objects as 

involved in discussion of resolv- 
ing power. 

Whether any two neighboring points on an object can be 
seen separated in the image given by any compound microscope 
depends on the objective alone. If they are not separated in the 
image formed by the objective, no amount of additional magni- 
fication by a projecting lens or by eye-pieces will ever succeed 
in separating them. This is illustrated in the series of photo- 
graphs in the plate on page 38. Fig. A is the photograph of the 
image of a diatom shell with an initial magnification X240; Figs. 
C and E are successive enlargements of A, i.e., X480 and X960 
respectively. It is quite apparent that little really new comes out 
with the successive enlargements. On the other hand if different 
objectives are used, which give successively better and better 



resolving power, we obtain the results shown on page 38, Figs. 
D, F and G. The actual magnifications in the latter series of 
figures are X420, X900 and X900 respectively. Fig B has a 
magnification about X 100, and the smallest resolving power 

We may now ask, "What is the closest distance by which 
two points can be separated and still appear as separate in the 
image?" K ' 

The Phenomenon of Diffraction 

In order to answer the last question it is necessary for us to 
speak of the nature of light. Up to the present we have been con- 
tent to represent rays of light by means of straight lines and to 
follow the direction of the light by the path indicated by these 


Fie. 24. 

How light spreads on passing 
through a small hole in a screen. 


I :••;•". Jr 

lines or rays. In other words we have tacitly assumed that light 
travels along straight lines in any homogeneous medium as though 
it consisted of luminous particles shot out from the source. This 
theory of light is known as the corpuscular theory. If we fol- 
lowed this reasoning, we would judge that there could be no 
limit to the minuteness of detail in any object which we might 
be able to resolve and magnify, as long as the light particles are 
assumed to be infinitely small. 

We have already applied the principle of the corpuscular 
theory m the presentation of the simplest image formation by 



means of a single lens (Fig. 5). In this figure we represented an 
ideal point, I, on the image, as corresponding to an ideal point, O, 
on the object. This linear representation, which connects object 
points with image points by straight lines, does not always give 
a true picture of observed effects. 

If we let a wide beam of parallel light rays fall upon a screen 
(AB, Fig. 24a) which has a small hole at P, we should expect 
a bright beam to emerge from P in the direction OPI, in such a 
manner that an observer's eye placed above or below the line PI 
would not receive any light from O. Actually this is not the 
case. An observer placed above or below PI does receive light 
from O (Fig. 24b). The orifice, P, behaves as though it had 
become a source of light itself, sending out a luminous cone 
toward the right of the screen. 

This is an example of the phenomenon called diffraction — an 
expression which means that light has the property of being able 
to bend around corners or obstacles and so does not travel exactly 
in straight lines throughout its course in any one medium, for 
example, in air. 

This phenomenon of diffraction of /light has a very important 
effect on the formation of images by lenses. If a small hole in a 
screen, S x , is illuminated from the left (Fig. 25) and used as an 
object point for a lens, L, the image formed at the point, I, on a 
second screen, S 2 , will not be merely a sharp point of light but a 

Fig. 25. The spurious disk or anti-point. 



regular pattern as shown— a bright central disk surrounded by 
alternate dark and bright rings. This experiment is best carried 
out by using an aperture stop, A, so that only the central part of 
the lens is used. If one uses a telescope to view the image given 
by the lens, L, quite a number of rings can be observed and easily 
photographed. Such a ring image of a point source came to be 
known to microscopists as a spurious disk or anti-point. By no 
amount of manipulation can one get rid of this departure from 
an ideal image point. 

If, instead of considering the object point as a bright point 
of light, we look at a small particle through a system of lenses 
we observe in the image the converse of the image shown on the 
screen in Fig. 25, that is, a dark central disk surrounded by a series 
of alternate bright and dark bands. Now if we imagine a second 
small particle at O' very near to the original particle at O, and 
just above it, the image pattern of the two points is as depicted 
in Fig. 26. The spurious disk images may overlap to such an 


Fie. 26. Overlapping of the images of two neighboring object points. 

extent that it becomes difficult to determine that there really are 
two particles in view. In such a case we say that the two particles 
in the object are not resolved. The critical problem is, "How 
close together can O and O' be and still be distinguishable as two 
separate points in the image?" 



Courtesy University of Toronto 

Tip of a cat's whisker magnified 4,600 times. 

Courtesy RCA Laboratories 

Face powder particle magnified 47,800 times. 



' """Mi 



L '■:>••'' 



|"'',vT,; r 

•:. :x£sx. 
- .- 




' f 


Pfev ',' 





Explanation of plate on page 38 

The plate is arranged to show the relation of magnification to resolving 
power, which depends on the numerical aperture (N.A.) of a microscope. 
All the pictures are of portions of the shell of a species of diatom, Pleuro- 
sigma angulatum. 

A to G are optical (light) microscope pictures. 

H and I are electron microscope pictures. 

B is taken with N.A. equal to 0.5, magnification X 100. 

A, C, and E were each taken with the N.A. equal to 0.75 but with 
magnifications X 240, X 480, and X 960, respectively; there is not much 
increase in detail visible as the magnification increases. 

D, F, and G were taken with progressively greater N.A., viz., 0.85, 
1.00, and 1.25, respectively, and with magnifications: X420, X900, and 
X 900. The detail is progressively better as the N.A. is increased. 

H is an electron microscope picture with magnification about X 5,000. 
This shows the real position of the holes in the shell. I is a broken edge 
of the shell of about the same magnification as H. 

The optical microscope pictures were taken by Dr. D. H. Hamly of 
the Department of Botany, University of Toronto. 



" ^.,^ -'■, : .'*'..1 > V'-T">*'. '. > 




Explanation of plate on page 40 

This plate reproduces pictures of various kinds of diatoms. 

A is an optical photograph (by Dr. L. T. Newman) of a selected speci- 
men of the type Synedra delicatissima kindly supplied by Dr. Paul S. 
Conger of Washington, D. C. Note that there is an odd circular type 
of diatom quite foreign to the others. The magnification is X 2,000. 

B is a made-up slide showing a random collection of electron micro- 
scope pictures of various types of diatoms found in a random sample 
of diatomaceous earth. The elongated ladder-like structures are similar 
to those types shown in A. 

C is an electron microscope picture of the diatom type, Cyclotella sp., 
magnification X 25,000. 

D is an electron microscope picture of the diatom type, Cyclotella sp., 
magnification, X 5,000. 

Courtesy American Cyanamid Co. 

(Above) Organic Dye Crystals, X 21,000. The original magnification of the electron 
micrograph was X 7,000 and it was enlarged to an additional magnification of X3, 

(Below) Kraft Pulp from Paper Beater, X 10,500. The original magnification of the electron 
micrograph was X 3,500, enlarged X3. 

Chapter 3 

The Meaning of a Scientific Theory 

In the preceding chapter we have given some answers to the 
question, "What does light do?" The present chapter will deal 
with the more difficult question, "What is light?" 

The mere act of formulating these two questions suggests the 
existence of two phases of scientific development. In the first 
phase man has learned how to make use of the forces of nature 
for his own comfort and satisfaction; he may be content with 
this. But there have always lived some men who have tried to 
answer the second question, "What is the natural force that I am 
making use of?" So we find in the study of science, on the one 
hand the engineering trend, and on the other the theorizing and 
philosophical trend. 

The formation of a theory, which is the answer to the latter 
question, requires a critical review of all known experimental facts 
relating to the particular natural phenomenon. The acceptability 
of any theory depends on whether or not We are satisfied in our 
minds with the picture we form to explain the facts. 

However, we should first inquire, "What do we mean by 
a scientific theory?" 

We have many natural phenomena to theorize about. What 
is heat? What is sound? What is light? What is gravitation? 
In each case we try to conceive in our minds a mechanical model 
which will produce the observed effect. 

What do we mean by a mechanical model? We play around 
with various machines and gadgets which do things for us. We 
know the forces involved in springs; we experience the pulls in 
a tug-of-war; we are familiar with the flow of liquids through 
pipes; the intricate motions of tennis balls, baseballs, and golf balls 



are matters of everyday experience. So when we come to account 
for such an intangible, nebulous thing as, for example, heat, which 
we say "flows" from one place to another in material bodies, we 
try to imagine what is flowing. The earliest theory or model of 
heat was the picture of an imponderable, invisible, elusive fluid, 
known by the name, caloric, which was supposed to flow from a 
hot body to a colder body. No experiment was ever devised 
which could really make such a fluid evident to our senses. Of 
course, everyone now says, "Heat is a mode of motion," by which 
we mean that we think we have a more satisfactory mechanical 
model for heat; we say that "heat possessed by a piece of matter is 
the sum total of the energy of motion of the molecules, and 
when heat is transferred from one place to another, fast-moving 
particles are communicating their motion to slower-moving 
molecules and make the latter move more quickly." In our first 
rough picture we imagine that these particles act like swarms of 
minute tennis balls, colliding and bouncing back in a random 

Our judgment as to what is a true theory is merely our answer 
to the question, "Which is the most satisfactory mechanical 
model which will best foretell experimental results?" 

We cannot say that any scientific theory is true or false; we 
decide merely which mechanical model is experimentally the most 
satisfactory. This is the sense in which we shall speak of a theory. 

Theories of Light in Ancient Times 

During the earliest period of recorded human speculation 
about the nature of light very few optical facts were known. 
People saw with their eyes the things around them and a few 
inquisitive minds wondered how the process of vision came about. 
The early Greek philosophers attempted explanations of the 
phenomenon of light, but their experimental knowledge was con- 
fined to three facts (1) transmission of light in straight lines, 
(2) reflection from smooth polished surfaces, and (3) bending of 
light by refraction in passing from one medium to another. 

° Pythagoras (582-497 B.C.) taught that vision was caused by 
particles continually projected from the surfaces of visible objects 


into the pupil of the eye. The teaching of Plato (427-347 B.C.) 
was rather more elaborate; he held that vision was brought about 
by the union or interaction of something emitted from the eye 
itself and something else emanating from the object. Plato's 
model was threefold; a stream of divine fire emitted by the eye 
became united with an influence coming from the sun and this 
combination reacted with a third emanation from the object 
and so accomplished vision. The simpler Pythagorean idea sur- 
vives today in what is known as the corpuscular theory of light. 
Aristotle (384-322 B.C.) discarded these emission theories of 
light and substituted for them what might be looked upon as a 
nebulous forecast of the modern wave theory, he maintained that 
light was merely a quality of, or action in, a space-filling medium 
which he called the pellucid — the forerunner of the modern ether. 

Modern Theories: Corpuscular Theory and Wave Theory 

Little would be gained by trying to unravel the intricacies of 
the theories regarding light down through the ages. The out- 
burst of experimental science beginning in the sixteenth century 
revealed in the course of one century many of the fundamental 
properties of light and laid the foundations of all future work 
in this field. 

In 1608 the first telescope of which there is a published 
account was constructed by a Belgian spectacle-maker, Hans 
Lippershey. One year later, using this as a starting point, Galileo 
built a telescope of such perfection that he was able to see, for the 
first time in human experience, the satellites of Jupiter; this opened 
a new field to the astronomer. 

In 1621, Willebrod Snell, a Dutch scientist, discovered the 
laws governing the refraction of light, the phenomenon illus- 
trated in Figs. 2 to 5. These laws enable us to foretell the amount 
of bending which a light ray undergoes as it passes from one 
medium to another. 

In 1675, Olaf Roemer, a Dane, discovered that light is propa- 
gated through space at a finite, though very large, velocity and 
not instantaneously, as previously supposed. He made this dis- 
covery by observing that the times of the eclipses of the satellites 


of Jupiter did not show the regularity that one would have 
expected. He accounted for this irregularity by assuming that 
the differences were due to the fact that the light took a longer 
or shorter time to come from Jupiter to the earth as the distance 
between these two planets changed due to their individual rota- 
tions about the sun (Fig. 27). He found that this anomaly was 
completely accounted for if one assumes that light travels through 
space at the rate of 192,000 miles per second— the first experi- 
mental determination of this important quantity. 


January 1st s\~*j' 

July 1st 

/ \ \ Distance light travels 

/ \ \ from Jupiter to Earth 

/ Distance light \ * yt /r January 1st 

l travels fro/T7-~~\ *\ -^ 

/ Jupiter to Earth W''' \ N N 

Fig. 27. ' July 1st V 

I i firth V \ 

Roemer's determina- J / u ^' 

tion of the velocity of ' \ 

!l ght- I \ Sun 

«JD Earth 

" January 1st 



About the same time (1672) Sir Isaac Newton made the 
remarkable discovery that ordinary white light, such as sunlight 
or light from an incandescent solid, is broken up by a prism into 
many colored components, giving the colors of the so-called 
spectrum— a distribution of colors already familiar in the rain- 
bow; in fact this discovery of Newton's at once offered an expla- 
nation of the formation of rainbows. 

To complete the contributions of this wonderful century of 
progress we record that in 1690 Christian Huyghens, a Hol- 
lander, observed that the natural crystal, Iceland spar, could 
break up a single beam of light sent through it into two parts, 
which travelled through the crystal in different directions. This 
discovery revealed the phenomenon which we call polarization 
of light, a property of light which has become a matter of every- 



day knowledge since the invention of Polaroid. This material 
consists of a very thin film of an organic substance made up of 
long-chain molecules. The film is mounted in a frame on a glass 
disk. By a special process the molecules are lined up like fibre 
layers with their long axes parallel to one another. The effect 
of polarization can be demonstrated by passing light through two 
of these disks in succession. If the molecules in the first disk 
are parallel to the molecules in the second, most of the light 
will pass through both disks. If the second disk is then turned 
in its own plane through 90° with respect to the first disk, no 
light will emerge from the second disk. A further rotation of 
90° in the same direction will then allow the light to pass through 
as before. 

Vibrations in 
random planes 

(a) parallel 

Fig. 28a. Polaroid crystals parallel in P and A. 

Vtb rations in 
random planes 

V/brations //7 


lYo //ght 

Crystals rotated to 
lie at right angles to 
those In P 

Fig. 28b. Polaroid crystals in A at right angles to those in P. 

The construction of a theory which would be able to explain 
satisfactorily all these phenomena was a much more difficult task 


than that which had confronted the Greek philosophers. It is 
small wonder then that many different theories, upheld by differ- 
ent schools of thought, co-existed for long periods of time. 

In Huyghens' and Newton's time two opposing theories 
existed; these were known as the wave theory of Huyghens and 
the corpuscular theory of Newton. On close examination of New- 
ton's own statements it will be found that his theory was the more 
general one and that it left room for both the corpuscular and 
the wave concept. 

In so far as light is a transmission of energy through space we 
would be justified in accepting a corpuscular theory, according 
to which a light source shoots out in all directions fine particles 
or corpuscles which enter our eyes and affect the retina (Fig. 29). 



Fig. 29. The transmission of energy through space by a moving mass. 

But there is a way in which energy can be transmitted 
through space without the actual transmission through space of 
any particles or corpuscles. One of the experiments in physics 
which almost every child performs illustrates this fact (Fig. 30). 
A row of dominoes, each standing on end and near its neighbor, 
can be arranged in any fantastic figure and, when one end 
domino is toppled over, a pulse of falling dominoes passes along 
the line. The motion, and therefore the energy, is transmitted 
through space, but matter is not transferred from the starting 



l/ounq physic/stdtu/ork 

Fig. 30. The transmission of energy through space without the transfer of any mass. 

point to the end point. Similar, but more far-reaching, phe- 
nomena are observed in the waves on the surface of water. These 
waves are caused by the winds disturbing the equilibrium of a 
quiet, flat surface, and we know from the destructiveness of 
violent storms that these waves carry along enormous amounts 
of energy. But if we toss a stick on the surface at any point we 
see that it is not carried along by the wave motion but merely 
bobs up and down close to the point where it was thrown in. 
From this we gather that the water itself is not carried along with 
the wave motion, and that consequently, energy alone must be 

We may say then that, as far as light is merely a transmission 
of energy, we would be justified in accepting a wave theory: that 
a light source initiates a vibrational motion in some medium, and 
that energy is carried along without the passage of material par- 
ticles along the line of the motion. 

Although we have outlined above the approaches to a solu- 
tion we are still not very convinced that we can answer the ques- 
tion, "What is Light?" 



x woo 



Explanation of plate on page 50 

This plate also shows the comparison between pictures taken with the 
light microscope and the electron microscope. 

A. Normal picture of a new razor blade. 

B. Light microscope picture of small portion of the same with magni- 
fication X315. 

C. Electron microscope picture of a small portion of B with total mag- 
nification about X 5,000. 

The width of the small line above A represents the total section in- 
cluded in B and the thick block over B represents the total width of the 
part of B included in C. 

D is an optical picture of asbestos powder ( X 1,100). 

E is the electron picture ( X 18,000) of the portion outlined by the 
small inked-in square area in D. The heavy bar in E is the hazy line 
through the square in D. 

These are good illustrations of the fact that the electron microscope 
reveals structures in a sample entirely unseen in the light microscope. 

(Above) 2-1/x Fraction of Wyoming Bentonite, X42,000. Courtesy A. Prebus, Ohio State 

C iS!c* Ka ° lin CIay ' X21, ° 00 ' (7 ' O00X electronic » 3X °P tica1 )- Courtesy American 

Chapter 4 


Is Light a Wave Motion? 

Probably the best way to unfold the present state of the theory 
of any natural phenomenon is to follow the historical develop- 
ment, proceeding step by step as the human race has done. As 
for light, we have seen that from the earliest times men have 
been aware of the phenomena of reflection and refraction and 
have accepted the idea that light travels in straight lines in any 
homogeneous medium. 

The first experimental result which disturbed this concept 
was the discovery of diffraction, as illustrated in Fig. 24. This 
effect could not be explained by the corpuscular theory; we can- 
not conceive of particles, shot off in a straight-line course, bend- 
ing around the corners of obstacles in their path. 

It is natural, when one is confronted with such an apparent 
deadlock, to review ideas that may have been proposed at an 
earlier stage and also to re-examine experiments in other domains 
which might yield a clue to a new approach to the explanation 
of the nature of the phenomenon in question. 

As for light, it had been the dream of natural philosophers, 
from Aristotle to Newton, that a medium of some sort filled all 
space. Various properties were assigned to this medium by 
different dreamers, all aimed at making it a carrier, in some way 
or other, of light energy from the source to the object finally 
illuminated. The concept of light corpuscles, to be sure, did 
not require such a medium, but it failed in the face of the demon- 
strable facts of diffraction. 

Transmission of energy by a wave motion in a material 
medium suggested itself, since this phenomenon was already 
familiar to everyone in the action of water waves. The question 




then was raised, "Is such a wave concept able to explain diffrac- 
tion?" Let us therefore study the behavior of water waves and 
see what can be learned from them that might be usefully applied 
to a model for light waves. 

The Diffraction of Water Waves 

The experiment depicted in Fig. 24 for the phenomenon of 
light can easily be duplicated with water waves. If the bottom 
of a large flat pan, Fig. 31, is just covered with water and two 
flat metal bars are placed so as to make a barrier, AB, across the 
water surface with the exception of a narrow opening at P, the 
experiment corresponding to Fig. 24 for light can be performed 
by means of waves upon the surface of the water. Drop a little 
pebble into the water, say at X; the wavelets will spread out from 
X as a center and will be reflected back from AB except at the 
opening, P, where some of the motion gets through to the other 
side. But instead of passing along a narrow path as indicated by 
the line PQ, the motion that gets through P spreads out all around 
the surface in the form of little wavelets which seen to emanate 
from the point P as center. 

Fig. 31. 

How a wave motion spreads 
on the surface of water after 
passing through a small opening 
in a bounding wall. 

It is on account of analogies such as this that scientists were 
constrained to say that light acted as though it had the proper- 
ties of a wave motion in some medium. 


The crucial property of wave motion, which is suggested by 
the phenomenon of diffraction, is that apparently every point 
on a wave front can be looked upon as being able to act as a new 
center from which the wave motion can spread anew, as is shown 
at the point P in both Figs. 24 and 31. 

The Elements of Wave Motion 

Our fundamental ideas regarding the propagation of energy 
by wave motion come from the actual motion which we observe 
in waves on the surface of water. Fig. 32 illustrates what hap- 
pens when we drop a stone on the smooth surface of still water. 
As the stone strikes the surface, wavelets spread out in concentric 
circles from the point of contact as center, and a floating object 
on the surface at any point will bob up and down. Fig. 32 may 
be taken to represent graphically two facts about this wave 
motion. First, if we imagine that the surface could be kept 

f\ \ A f\ 


' 3"— XsJ 



Fig. 32. Graph of water wave showing the meaning of wave length, phase, fre- 
quency and velocity for a typical wave motion. 

frozen at any given instant, the curve shows the relative position 
of the surface points at that instant; in the second place, if we 
imagine the curve travelling along in the direction of A to B, for 
example, with the velocity of the wave itself, we have a picture 
of the progress of the motion; and, if we fix our mind on any 
one point, as Y x , we get a true idea of the motion of a water 
particle at this point as the wave sweeps along. Since the actual 
motion of the particle is at right angles to the direction in which 
the wave itself is progressing, this is said to be a transverse wave. 
We can learn from this figure some of the technical language 
applied to wave motion in general. The distance from crest to 
crest or from trough to trough is called a wave length; two points 
which are in corresponding positions in the crests or troughs, as 



Xi and X 2 are said to be in the same phase. The number of 
times the particle at a point on the surface bobs up and down in 
one second, or, as we might say, the number of times a particle 
completes a cycle in a second, is known as the frequency of the 
wave motion. It is at once apparent that the velocity with which 
the disturbance moves forward is equal to the number of wave 
lengths marked out per second, i.e., is equal to the frequency 
multiplied by the wave length. This is expressed in mathematical 
shorthand by the relation: 

v = vk 

where v means velocity, v frequency and X wave length. 

Fig. 33 represents the same phenomenon as Fig. 32, but 
shows the concentric circles as we look down on the surface; a 
curve similar to Fig. 32 is repeated below. This illustration is 
extended in Fig. 34 to show the interaction of two independent 
sets of concentric ripples. 

If we suppose that two stones are dropped in at the same 
time at points O' and O", two sets of wavelets would develop 

Fig. 33. FlG> 34> 

The superposition of two systems of waves on the surface of water. 

from these points as centers. These two would overlap as they 
proceed; Fig. 34 represents an instantaneous picture of a section 
of the surface through the points O' and O". The surface par- 
ticles at any point, P, will have assumed the displacement due to 


the sum of the two motions, and so the real position of the surface 
points at the particular instant chosen will be that given by the 
dotted curve. At the position F, for example, the surface point 
will be at its original position of rest; that is, at this point the two 
wave motions combine to give zero displacement. Again, at 
the position P, there will be an abnormally large displacement 
due to the two disturbances adding their effects at this particular 


This interaction of two independent wave motions is known 
technically as interference. 

From a study of Fig. 34 it will be apparent that, as the wave- 
lets progress, the resulting (dotted) curve will always cut the axis 
at the same points, i.e., those points at which the lines V ¥ inter- 
sect the axis. That is, at certain places on the surface of the water 
the two independent waves (from O' and O" respectively) will 
always neutralize each other; in other words, at certain points on 
the surface there will be no motion at any time. 

Courtesy RCA Laboratories 

Trachea of honey bee (X 15,000). 

Courtesy A. P rebus, Ohio State University 
(Above) California Montmorillonite, X 38,000. 
(Below) Fraction of Attapulgite (200-50^) , X 35,000. 

Chapter 5 

Wave Motion in Gases and Liquids 

Although we get our simplest example of wave motion from 
surface water waves, this phenomenon is quite inadequate to serve 
as a model of wave motions propagated through space in three 
dimensions. The water wave spreading over a surface is confined 
to that surface and cannot serve as a model for the motion caused 
by the explosion of a depth bomb below the surface, which 
spreads out from the source in all directions in the ocean, nor for 
the disturbance set up in the air by, say, a firecracker. These 
two are examples of wave motions in three dimensions; but in 
both cases we have energy being transmitted from one point to 
another in the form of sound. 

For a more detailed description of the phenomenon we illus- 
trate, in Fig. 35 (a, b, c), the propagation of the sound from a 
gong which has been struck by a drum-stick; this drawing was 
constructed in accordance with our accepted model of the air, 
namely, that it is a collection of small spherical particles swarming 
around in random motion. The metal disk vibrates when struck. 
As it moves to the right it compresses the air in its immediate 
vicinity to the right; then it swings back toward the left, making a 
compression to the left side and leaving a rarefaction on the right 
side. The air near the disk goes through this succession of oscilla- 
tions as long as the vibrations of the disk continue. 

The events taking place on the right side of the disk are 
shown diagrammatically in Fig. 35c which represents, as it were, 
a snapshot exposure of the air particles. The shaded regions are 
those of compression and the dotted regions are those of rare- 
faction; the little arrows indicate the direction in which the air 




particles have moved just before the snapshot was taken. A 
moment later the disk will bulge to the left and the state of affairs 
will be reversed in every respect. We can really measure the 
pressures along the line CO' as the sound is travelling, and the 
curve in Fig. 35b shows graphically the relation between the pres- 
sure, d, at a point along CO' and the distance, s, of the point from 
the center of the disk at the time of the snapshot. As the direction 
of motion of the air particle is the same as that in which the sound 
moves, this form of wave motion is said to be longitudinal. 



N x 



kfc.-.xV..-.*\ii-.\i. _. o 



Pressures higher than normal 


(b) d^ 

> S(cm; 

Fig. 35. The motion of air particles in the neighborhood of a vibrating gong. As 
the sound travels to the right, (b) represents how the pressure (d) varies at any 
point along CO', the distance s from the diaphragm. 

The human ear, which acts as a receiver, responds to the 
variations in air pressure, and the sensations produced physiologi- 
cally are interpreted as sounds of different loudness and pitch. 
Since sound cannot be transmitted in a vacuum, we must accept 
the conclusion that the air is the mechanical link between the 
sound source, the gong, and the receiver, the ear. 

That sound is a form of energy is readily understood if one 


recalls that windows may be broken by the impact of sound waves 
resulting from an explosion at some distance. Although the 
results are not so violent, energy is also necessary to produce the 
motion of our ear drums as they record sounds falling upon them. 
Sound travels not only through air (and other gases) but also 
through liquids and solids. In every case the origin of the sound 
is a periodic disturbance which causes a slight displacement of the 
molecules of a material medium, and the resulting periodic wave 
motion is due to the action of forces which always come into play 
to push these disturbed molecules back into their original posi- 
tions. The velocity with which the disturbance moves through 
any medium depends on the nature of the medium, and the molec- 
ular forces called into play to restore the molecules to their orig- 
inal positions are known as elastic forces. We call this property 
of a medium, elasticity, every medium has its own peculiar elastic 
properties which are indicated by so-called elastic constants. 

Sound waves in any medium are thus elastic waves. If the 
frequency of the motion set up in any medium is between 20 
cycles and 20,000 cycles per second, the resulting motion affects 
our ears, causing a sensation in the brain which we call sound. 
Any disturbance set up in a given medium, whether we can hear 
it as a sound or not, travels with the same kind of vibratory motion 
and with the same velocity. In fact, the velocity of any such dis- 
turbance in any given medium depends only on the elastic con- 
stant of that medium and its density, and not at all on the sound 
characteristics, such as pitch and intensity. 

Consequently, a disturbance caused by an explosion in air 
travels with the same velocity in still air as the sweetest note of 
a bird. Again, when a depth charge is set off in the ocean at a 
considerable distance from the surface, the disturbance travels out 
from the center of the explosion in spherical waves of compression 
and rarefaction with a velocity which is the same as the velocity 
of any ordinary sound in water, or about 4,500 feet per second. 
The direction of the to and fro motion of the molecules in liquids 
is always back and forth along the direction of progress of the 
disturbance, so that here, as in the case of gases, we have longi- 
tudinal wave motion. This is the only possible kind of wave 
motion that can exist in liquids and gases. 


Wave Motion in Solids 

When we come to study how sound travels in solids, we find 
that two kinds of sound waves are possible. If we pluck a violin 
string in the ordinary way we obtain a rich, pleasing note: but if, 
instead of plucking the string, we pull it with a resined cloth along 
its length, we obtain a screeching, unpleasant note. Both these 
notes are due to vibrations set up among the molecules of the 
string; but from the manner in which the disturbance is caused 
we see that the motion of the molecules, in the first case, must be 
perpendicular to the length of the string or transverse, whereas in 
the second case the motion will be longitudinal, i.e., parallel to the 
length of the string. So if we could see the motion of the mole- 
cules inside the string, we should observe that the disturbance is 
carried along by a transverse wave in one case and in the other by 
a longitudinal wave. Experiment has shown that these two wave 
motions in solids are regulated by two separate elastic constants, 
which correspond to two separate elastic properties of the solid of 
which the string is made. 

The Propagation of Waves in Matter 

In summarizing our various experiments demonstrating the 
propagation of different waves in material media we should search 
for a common criterion that makes the establishment of the waves 
possible in all cases. To begin with, the equilibrium of the 
medium in question is in every case disturbed by an external force 
which imparts to the particles of the medium a displacement in 
the direction of the force. This displacement is not permanent. 
The medium offers a resistance which tends to restore the equi- 
librium position of the particles when the external force is re- 
moved. Plastic solids, such as plasticine and dough, which do not 
possess these strong molecular forces, cannot maintain a wave 
motion and consequently cannot transmit sound. The motion of 
the diaphragm of the gong in Fig. 35 sets up longitudinal waves 
in the air due to the ability of the gong to vibrate for some time 
after being hit by the drum-stick. In this case we have not only 
the vibrations in the air but also a wave motion travelling to and 
fro within the metal of the plate at a definite rate — a wave in the 
solid itself. 


It is to these internal wave motions that we referred in pointing 
out the existence of wave motions in the material of the violin 
string. The disturbances which we call earthquakes are wave 
motions transmitted through the solid crust of the earth, and it 
is this kind of wave motion in a large volume of homogeneous 
solid that has really served as the prototype for the wave theory 
of light. 

The Velocity of a Wave Motion in Matter 

The interrelation between the elastic constants and the density 
of solids and the velocity of a wave motion was very fully ex- 
plored many years ago, beginning with the experiments of Robert 
Hooke (1635-1703). It was found that the velocity, v, was 
given by the relation 

v = S/E/d 

where E is the elastic constant and d the density of the solid. 

In Table 2 is collected some information regarding the ve- 
locity of sound in some typical media, as deduced from this rela- 
tion; in every case the numbers obtained for the velocities agree 
very well with those obtained from experiment. Of course all 
these quantities change with changing temperature; usually these 
measurements are taken at a standard room temperature, 20° 
centigrade or 68° Fahrenheit. 

Table 2. Values of the Velocity of Sound. 

Substance Elastic Constant Density Velocity of Sound* 

(dynes per (gram per cc.) (cm. (It. 

sq. cm.) " per sec.) per sec.) 

Aluminum 7.0 X 10" 2.70 510,400 16,740 

Copper 12.3 X 10 11 8.93 356,000 11,670 

L ea d 1-6 X 10 11 11.37 122,700 4,026 

Steel 20.9 X 10 11 7.8 499,000 16,360 

Water (at 20°C) 141,000 4,625 

Air(at0°C) 33,130 1,087 

Hydrogen (at 18°C) 130,100 4,268 

*From Smithsonian Tables. 


A Medium Necessary for Light Wave Motion 

Since the conception of a wave theory of light was based on 
experimental proofs of the existence of wave motions in material 
media, it was incumbent on the supporters of this idea to provide 
a model medium in which such a wave motion could take place. 
It was postulated that a satisfactory medium did exist through- 
out all space; this was given the name ether— just another of the 
invisible, imponderable fluids which have often served to satisfy 
the human longing for an explanation of mysterious natural 

Manifestly this new medium must have an elastic constant and 
a density such that the velocity of light would be given by the 


When first suggested, the ether was assumed to be a fluid, i.e., a 
liquid or a gas, and so the motion of the particles of the ether was 
supposed to be longitudinal. 

Modern determinations of the velocity of light make it 186,- 
000 miles per second (or 30,000,000,000 cms. per sec.)— a per- 
fectly enormous velocity compared with any that had ever been 
observed for any sound or elastic wave. In order that such a ve- 
locity should exist in any medium, the value of the elastic constant, 
E, would have to be extremely large or the density of the medium 
would have to be extremely small, or both of these conditions 
would have to obtain at the same time to a lesser degree. In effect, 
if the ether were supposed to have an elastic constant, E, equal to 
that of steel, i.e., if it were as rigid as steel, the density could not 
be larger than about l/50th that of the lightest known substance, 
hydrogen gas; if, on the other hand, the ether were supposed to 
have a density equal to that of hydrogen, the elastic rigidity would 
have to be about 50 times that of steel, one of the stiffest materials. 




Prodigiosus bacillus under optical 
microscope ( X 2,000) . 

under elec- 
tron micro- 

( X 16,400.) 

This plate is also designed to compare light microscope and electron 
microscope pictures of the same specimen— the Bacillus Prodigiosus. The 
optical pictures were taken by Dr. J. Craigie of the Department of 
Hygiene, University of Toronto. The line marked lp gives the scale 
of the electron picture, this length being 1710,000th of a cm. or l/25,000th 
of an inch. It is quite apparent that the electron microscope may be 
counted on to show structure in such small bodies. It should be said that 
none of the electron microscope specimens are stained. 

Chapter 6 

Interference of Light 

Figs. 24 and 25 illustrate actual observations of how light acts 
and we have seen how the phenomenon presented in Fig. 24 may 
be exactly reproduced with water waves. 

We have now to show that the phenomenon of interference, 
demonstrated by water waves, is also an essential property of 
light, and consequently affords additional reasons for looking 
upon light as a wave motion. 

The simplest experiment showing interference of light is illus- 
trated in Fig. 36. Light from a source, S, is shielded by a screen 
with a narrow slit, P, which acts as a line source for the space 
to the right of the screen. The light from P spreads out and falls 
on a second screen which has two slits A and B, symmetrically 
placed with respect to the center line of the figure. The light 
passing through the slits, A and B, is allowed to fall on a third 
screen some distance away, so that the cones of light from A and 
B will overlap. Now, instead of the last screen being uniformly 
illuminated, we observe on the screen alternate bands of light and 

Fig. 36. How two light beams "interfere" with each other. The lines on the last 
screen are symbolic of darkness. 




darkness. That is, we have the curious phenomenon of light from 
two sources, A and B, travelling independently to a screen and, 
with no obstacle involved, producing darkness at some points on 
the screen. By no stretch of the imagination can we convince 
ourselves that corpuscles travelling along can produce this effect. 
Fundamentally we must satisfy ourselves that two beams of light, 
each of which alone would give uniformly bright illumination of 
a screen, can so interact as to produce bands of light and darkness 
on the same screen when they are both shining on the screen at 
the same time. 


Fig. 37. The interference of two wave motions having a difference in phase of 

one-half a wave length. 

The wave theory gives a very natural forecast of such a result. 
Suppose we represent the distribution of compressions and rare- 
factions by curves as (a) in Fig. 37, similar to the wave motion 
curves of other figures. Let us consider the effects of combining 
two wave motions represented by Figs. 37a and 37b. If one wave 
crest tends to produce a compression of the ether and the trough 
of a second wave tends to produce a rarefaction at the same point 


simultaneously, the two forces, if equal in intensity, will cancel 
each other and the ether will remain undisturbed at that point. 
If this happens all along the path of two co-existing waves, which 
thus we describe as opposite in phase, then the ether will remain 
undisturbed all along the wave path. The effect is the same as if 
no waves had been propagated at all, or in other words we have 
two independent waves of light adding up to darkness. 

Fig. 38. The interference of two wave motions having a difference in phase of 
one-quarter of a wave length. 

Figures 37a-d are snapshot exposures of the distribution of 
compressions and rarefactions along the direction, s, in which the 
light is travelling. Wave 1 starts with a compression, wave 2 with 
a rarefaction, while both have the same wave length. It is ap- 
parent that one wave is a half wave length behind the other, or as 
we say technically, it is out of phase with the other by that 
amount. This leads to the complete cancellation of the two, if 


the two waves are superimposed as in Fig. 37c ; the next figure, 
37d, gives the graphical result, i.e., no action at any point along 
the path. As each wave represents an independent light beam of 
the same intensity in both cases, we must conclude that light plus 
light adds up to darkness under the particular circumstances that 
were just described. If, on the other hand, we arrange it so that 
wave 2 in Fig. 37b is shifted with respect to the first wave by a 
quarter of a wave length, we produce the conditions illustrated 
by Fig. 38 (a and b). Both waves have again the same wave 
length; they are shown superposed in Fig. 38c. When we com- 
bine these two so as to represent the distribution of compressions 
and rarefactions which is the resultant of the two, we obtain 
Fig. 38d, which is a new wave of different shape but still of the 
same wave length. Comparing Figs. 38d and 37d we realize that 
now the two waves do not cancel out all along the path from O 
to F as before, but only at the points B, D, and F. Between these 
points the medium is definitely disturbed and energy is propa- 
gated. If we were to expose instantaneously a strip of film along 
OF, this film, when developed, would show the pattern of 
Fig. 38e, i.e., dark regions at B. D, and F, and light regions 
between these points with shaded transitions. This example ex- 
plains how nicely the wave theory accounts for the phenomenon 
of interference, without any artificial assumptions. 

Acceptance of the Wave Theory 

By the year 1800 the wave motion theory of light was ac- 
cepted almost unanimously by scientists. The convincing ex- 
planation of many optical effects, such as reflection, refraction, 
diffraction, and interference, seemed to afford ample justification 
for ( . accepting the concept of a wave motion of the "ether" as a 
satisfactory theory. The explanation of the straight line propa- 
gation of light based on the wave model left something to be de- 
sired, and the nature of the ether itself, with such unique proper- 
ties as it was found necessary to endow it, remained an enigma. 

When the wave theory was first generally accepted, the ether 
was considered to be of the nature of a gas, and consequently the 
wave motion in the ether was supposed to resemble the wave 


motion in a gas. Now the only kind of wave motion that can 
exist in a gas (or a liquid) is one where the oscillation of the 
particles of the medium is longitudinal; that is, the actual to-and- 
fro motion of the oscillating particles is along the direction of 
transmission of the disturbance. However, if we refer back to 
Fig. 28, which illustrates the phenomenon of polarization of light, 
we realize that polarization can be explained satisfactorily if we 
imagine the vibrations to be transverse. But we cannot imagine 
the molecular arrangement in the Polaroid able to cut out longi- 
tudinal vibrations. Now transverse vibrations are possible only 
in solids; consequently, physicists were forced to look upon the 
ether as having the properties of a solid. We find a great deal 
written about the elastic solid theory of the ether during the last 

The Limit of Resolution of the Microscope 

With the foregoing background regarding light as a wave 
motion, we may now return to the question at the close of Chap- 
ter 2: "How close together can two object points be and still be 
distinguishable as two separate points in the image?" 

Reference to Fig. 25 will recall that any point source will give 
an image consisting of a system of dark and light rings, due to the 
diffraction of light — the so-called spurious disk. This set of 
rings is due to interference between portions of light which are 
transmitted by various parts of the lens. As was pointed out 
previously, if we have two objects, such as P and Q (Fig. 39), we 

i—-""* < / \ 

l^>,, :r i-.-u.u :: m 


, . M)„. 

Fig. 39. Lord Rayleigh's determination of the resolving power of a lens system. 


cannot separate them in the image if the two spurious disks, F and 
Q', overlap too much. 

Lord Rayleigh used this picture to deduce an answer to the 
above question. His argument applies also to complex lens com- 
binations, but in Fig. 39 we shall let the single lens, L/L", repre- 
sent diagrammatically the objective of a microscope. We shall 
state Rayleigh's result here and indicate how it is arrived at in a 
footnote which may be ignored by the non-technical reader. 

If d is the smallest value of PQ as defined above, Rayleigh 
deduced the formula: 

d = 


where A, is the wave length of the light used with the microscope 
and N.A. is the numerical aperture of the microscope.* The 
numerical aperture is a very important constant of a microscope 


setup; it is equal to the product: nX~-—, (Fig. 39), where n is 

the index of refraction of the medium between the object, PQ, 
and the first surface of the lens system. In mathematical parlance 

the fraction — — is called the sine of the angle A, and is written, 

sin A. It is common practice to write the capital letters, N.A. 
for n sin A. 

* The crux of Rayleigh's argument is that separation of the image points depends 
on how much light from Q ultimately arrives at the center of P\ (Note that we 
say P' here and not Q'.) Whether this light from Q is enough to cause trouble at 
P' or not depends on the differences between the distances travelled from Q to the 
center of P' by the portions of light from Q which pass through the various parts 
of the lens, L'L". The extreme difference of these paths is given by the difference 
between the paths QL'P' and QL"P'. Rayleigh assumed that, if this extreme differ 
ence of path equalled the wave length, X, of the light used, the interference of light 
from Q with the image of P at P' might be neglected. 

Since L'P' = L"P', the difference QL'P'-QL"P' will be just the difference QL'-QL". 
By a very simple calculation, keeping in mind that the distance PQ is very small 
and, consequently, that the angles L'PL" and L'QL" are almost equal, it follows that 
2PQ sin A = \, the wave length of the light in the region between PQ and the lens. 
From this we have that: 


2n sin A 

where \ is the value of the wave length of the light in a vacuum and n is the index 
of refraction of the medium between PO and the lens surface. 


According to this formula, the power to distinguish two points , 
which are very close together in an object is limited in practice 
by the value of the wave length of the light 1 used and by the value 
of the quantity N.A. The smaller the wave length of the light 
used, the smaller the limiting distance PQ may be; the larger the 
numerical aperture, N.A., the closer together the points P and Q 
may be. In the plate on page 38 are examples of microphoto- 
graphs of the same diatom with the following values of N.A.: 
for B— .5; for A, C and E— .75; for D— .85; for F— 1.00; for 
G — 1.25. It is quite apparent that detail is progressively better 
as the value of N.A. increases. 

The maximum value of N.A. for any lens system used in air is 
slightly less than unity; by using oil immersion objectives, for 
which n is greater than one, N.A. can be made nearly equal to 
1.5. Consequently the best we can do is to have 

d = i% 

Thus the actual value of d depends on the wave length of the 
light used. In Table 3 we give the wave lengths corresponding 
to the range of colors in the spectrum of white light. These 
dimensions are usually given in centimeters but, since the numbers 
turn out to be such small fractions of a centimeter, it has been 
found convenient to introduce smaller units of measurement. As 
several of these smaller units are in use it is necessary to keep them 
carefully differentiated. The basic relations between these other 
units and the fundamental unit, the centimeter, are as follows: 

1 millimeter (mm.) =0.1 cm. 

1 micron ({a) =0.001 mm. = 10" 3 mm. = 10" 4 cm. 

1 millimicron (mp) = 0.001 ^ = 10" 6 mm.^10" 7 cm. 

1 Angstrom unit (A.U.) =0.1 m{x=10- 7 mm. = 10" 8 cm. 

The range of wave lengths in the visible spectrum extends 
roughly from 4xl0~ 5 cm to 8XlO" 5 cm or, expressed differently 
from 4000 to 8000 A.U. 

If we select for purposes of calculation the wave length 
6X10" 5 cm, the smallest possible distance, d, for visible observa- 



Table 3. Wave Lengths of Light of Different Colors. 

Color range 

In Centimeters 

In Angstrom Units 


Above 7.70 XlO" 5 

Above 7700 


6.47-7.70 " 



5.88-6.47 " 



5.50-5.88 " 



4.92-5.50 " 



4.55-4.92 " 



3.60-4.55 " 



Below 3.60 " 

Below 3600 

tion is approximately 2 XlO" 5 cm, i.e., l/50,000th cm, or 
l/125,000th of an inch. This is then also the diameter of the 
smallest particle that we can see or photograph with the aid of 
the most powerful microscope, using ordinary visible light. 

If we make use of ultraviolet light and photography, we can 
take pictures of particles about one-half this size and so have 
visible in the photograph greater detail. 

This marks the impasse reached by the year 1900. 

Resolving Power and Magnification 

The argument presented in the preceding section tells us that 
with the help of even the finest microscope, the smallest particle 
that we can see must have linear dimensions of at least l/125,000th 
of an inch. No amount of magnification or re-magnification can 
reduce this minimum value. 

Since the naked eye cannot see differences in dimensions 
smaller than l/250th of an inch, it is at once apparent that a 
dimension of l/125,000th of an inch must be magnified at least 
500 diameters. This is usually written X 500. Though the eye 
can see very small detail, eye strain and fatigue are avoided by 
increasing this magnification from 2 to 6 times. This increase in 
magnification is dependent upon the initial contrast between 
object and field which, in the case of small transparent particles, 
is very low. When the contrast is low no increase in magnifica- 
tion will increase the amount of detail, since image contrast de- 
creases with increase in magnification. Magnifications of X 1200 


to X1800 are practical with well-stained materials and a good 
oil-immersion objective, but higher magnifications up to about 
X5000 are possible only with metallurgical materials showing 
exceptional contrast. 

Contrast is a primary requirement in the visual perception of 
any image. When contrast is low it can be increased optically by 
the use of suitable filters and photographically by the proper use 
of emulsions and developers. Enlargement reveals very small 
detail by bringing it to the size perceptible to the eye. Empty 
magnifications and low contrast result when the limits already 
mentioned are exceeded. 

The detail seen in unstained materials, such as cleaned diatoms, 
can be improved by increasing the contrast with the use of media 
of higher refractive index than that of cedar oil. Where the use 
of ultraviolet light is possible with unstained organic materials, 
such as thin cross-sections of tissues and of bacteria, the peculi- 
arities of differential fluorescence and absorption increase the 
contrast. While these special techniques have been very success- 
ful in revealing detail up to the limits of resolution, in ordinary 
practice these limits are very seldom attained. 

The Ultramicroscope 

About 1900 a completely new technique in microscopy was 
developed by Zsigmondy and Siedentopf of the Zeiss Company. 
Like many other scientific advances it involved the application 
of an ordinary well-known experimental fact in a new field. 

We are quite well aware that the air of any ordinary room is 
filled with motes and dust particles which are invisible when the 
room is well lighted. However, if we darken the room as a 
whole, admit a narrow beam of sunlight or throw a narrow beam 
from a projection lantern and view this bright beam from the 
side, we are made aware of the presence of thousands of motes or 
dust particles floating in the air. But if we look in and along the 
beam toward the source, the same dust particles or motes are 
quite invisible. 

The reason for this phenomenon is that the small motes or par- 
ticles scatter or diffuse or reflect in all directions the light which 






JK # 

• ijlpr - 5 


Explanation of plate on page 76 

The purpose of these pictures is to show the manner in which the 
electron microscope contributes to the intepretation of the best optical 
microscopic pictures. 

The objects are diatoms of a type often used for testing first class oil- 
immersion objectives. 

A shows photographs of Synedra delicatissima. For the oil-immersion 
pictures, perfectly symmetrical Kohler illumination and an aperture giv- 
ing N.A. 1.4 were used. The optical microscope (left) indicates the 
presence of rows of holes, but the electron microscope ( X 5,000) (right) 
goes further than this and shows positively the size and arrangement of 
these rows and holes. The holes are approximately 0.5/* in diameter, 
they are separated in the row, center to center, by a distance of 0.20/*; 
the rows themselves are approximately 0.90/x apart. 

B shows corresponding photographs of the diatom, Amphipleura pellu- 
cida. The optical pictures (left) were taken with the same Kohler 
illumination and the same N.A. 1.4. The optical microscope shows bars 
in the valves and hints at the presence of openings within the slots. The 
electron microscope (x 5,000) (right) shows clearly the small holes, 
approximately 0.14/* in diameter, with centers separated by 0.20/x. The 
rows are about 0.20/* apart. 

C shows a valve fragment of the well-known test diatom Pleurosigma 
angulatum photographed with symmetrical illumination and an aperture 
of N.A. 1.0. The magnification is enlarged optically to X 5,000. 

D is an electron picture of a portion of C with magnification X 5,000. 
That D itself does not tell the whole story given by the electron micro- 
scope is shown by the fact that E, a portion of D with total magnification 
X 15,000, shows that each individual hole in the shell is really a series of 
four holes approximately 0.1 /* in diameter separated from each other by 
0.03/*. The holes appearing at lower magnification are about 0.7/* apart. 

The optical pictures were made by Dr. D. H. Hamly of the Depart- 
ment of Botany, University of Toronto. 



falls on them. When we view the beam laterally our eyes catch 
the scattered light against a dark background, and so we say we 
see the particles. When the whole room is again illuminated the 
dark background disappears so that we can no longer distinguish 
the light scattered by the particles. 

Fig. 40 shows diagrammatically the essential features of the 
Zsigmondy-Siedentopf ultramicroscope. The sample of mate- 
rial containing the small particles — gold ruby glass or some other 



Fig. 40. The principle of the ultramicroscope. 

colloidal solution — is illuminated in one direction and viewed at 
right angles to the illuminating beam by means of an ordinary 
microscope. The 'ultra" part of this instrument is the unique 
illumination which enables particles to scatter light into the 
microscope so that they are seen against a dark background. 
One deficiency of such a set-up is that we are only made aware 
of the presence of such particles, but we cannot measure them or 
tell their shape. On the other hand, we can be made aware of 
the presence of particles very much smaller than those set by the 
limits given in the previous sections. The reason for this is that 
particles of any size will scatter or diffuse some light in all direc- 
tions, and whether we see such particles or not depends on 


whether the amount of scattered light which manages to enter 
the microscope, and consequently the eye of the observer, is 
sufficient to affect the retina. This is chiefly determined by the 
brightness of the illuminating beam. It has been calculated that 
we should be able to see ordinary molecules with the ultramicro- 
scope if we could illuminate them from a source as bright as the 
surface of the sun. 

Fixed 5tar 


Fig. 41. Planets are so near that a telescope shows details of their surfaces, and 
gives a distinct measure of their sizes; but fixed stars appear only as points of 
light, on account of their enormous distances from the earth. 

This phenomenon is analogous to one encountered in tele- 
scopic vision. The only heavenly bodies for which we can deter- 
mine the size by direct observation or of which we can see any 
detail are those belonging to our own solar family, such as planets 
and their satellites (Fig. 41). But we are made aware of the 
existence and position of the far-away fixed stars because they are 
so brightly illuminated that our telescopes gather and transmit to 
our eyes sufficient light to affect the retina. We are however 
not able to tell anything about the size and shape of these fixed 
stars by vision alone, no matter how huge our telescopes. 

While the development of the electron microscope has ren- 
dered the ultramicroscope outmoded as far as the determination 
of the size of small particles is concerned, this instrument is still 
very important as it affords the best method of observing the 
Brownian movement shown by colloidal particles. 



V» Car 


- - 

Courtesy Columbian Carbon Co. 

The electron microscope at the University of Toronto attacks its first problem in indus- 
trial research. Left to right, Professor E. F. Burton, Albert Prebus, and William Ladd. 

Chapter 7 


Faraday's Conception of Fields of Force 

Up to the present we have been speaking of light as an isolated 
phenomenon of nature. It is probably not too much to say that, 
as far as physical sciences are concerned, the most important dis- 
covery of the nineteenth century was that the phenomena of 
light, electricity and magnetism are intimately related. In fact 
it has been proven beyond the shadow of a doubt that light is 
itself an electromagnetic phenomenon. The twentieth century 
has already gone a great step further by showing that the relation 
between light and matter is so intimate that it is possible to bring 
about a transformation of light (radiation) into matter and of 
matter into light. Indeed, both light and matter present a dual- 
istic aspect — a wave motion and a corpuscular discreteness. In 
developing this idea we shall have first to recall the discoveries 
in fields of natural phenomena apart from but paralleling develop- 
ments regarding light. 

During the two hundred years preceding 1800 the funda- 
mental facts regarding electricity and magnetism were known, 
in so far as the existence of two kinds of electricity and two kinds 
of magnetic poles is concerned. At this time it was not known 
that electrical charges and magnets had the slightest connection. 
Following a suggestion by Benjamin Franklin, the charge assumed 
by ebonite when it is rubbed with fur was called negative electric- 
ity, and that assumed by glass when it is rubbed with silk was 
called positive electricity. When two pith balls, one touched to 
charged ebonite and the other touched to charged glass, were 
brought close together they attracted each other; but if the two 
were touched to the same electrically charged body (ebonite or 
glass) they repelled each other (Fig. 42) . 




Since there are forces existing between these charged pith 
balls, energy has been transmitted through space from one body 
to the other. The question arises, "How does this interaction of 
forces take place?" 

This problem will naturally bring to mind the similar one 
involved in the propagation of light. There is, however, a dif- 
ference between the two types of energy transfer. In the case of 
light, it is only a transfer of energy from a source to the receiver, 
which may be the human eye or a photoelectric cell or a thermo- 
couple. It is a unidirectional transfer of energy, a flow in one 
direction, however it comes about. This state of affairs does not 
prevail in the case of the two pith balls. 

fS SS/S//Y 



Uncharged Charged 

Fig. 42. How charged pith balls attract and repel each other. 

The repulsion or attraction in the case of the two balls involves 
a certain amount of work, so that energy must have been supplied. 
But we cannot say that ball A is the source and B the receiver of 
energy or vice versa. Again, unlike light, this phenomenon does 
not involve a continuous transfer of energy from source to re- 
ceiver, but the setting up of a state of equilibrium in which one 
pith ball is kept at a certain distance from the other in spite of the 
effect of gravity, which would make the balls hang vertically. 

Most of the early workers were content to cover up their 
ignorance of this mysterious state of affairs by saying that it was 
just a case of "action at a distance," quite analogous to the mutual 
attraction of the sun and the earth. 

An exactly similar situation developed with regard to the 
action of one magnet on another. In the minds of the early scien- 
tists, like poles repelled each other and unlike poles attracted each 


other because of some innate property of the magnets, not at all 
dependent on what is in the space between them. 

Faraday, whose scientific thinking was dominated by mental 
pictures of mechanical models of natural phenomena, was consti- 
tutionally a vigorous opponent of the ideas involved in the phrase 
"action at a distance." He believed firmly that the interaction 
between charged pith balls, as well as that between magnets, could 
be explained only by assuming that something was happening in 
the space surrounding the pith balls (or the magnets); Faraday 
held that there was set up in the intervening space a state of strain 
in some medium filling the space. The space around the reacting 
bodies thus became the mechanical link causing the repulsion or 
attraction. This state of affairs is expressed by saying that there is 
a field of force (electrical) around the charged pith balls and an- 
other different field of force (magnetic) surrounding the magnets. 

Media for Electrical and Magnetic Fields of Force 

These two media, electrical and magnetic, would have to be 
imagined to exist coincident with the ether which was assumed 
for the explanation of the propagation of light. 

The whole question was complicated by the discovery by 
Oersted in 1820 that there existed a magnetic field in the space 
about a circuit bearing an electric current, and the discovery made 
later by both Faraday and Henry that currents of electricity 
could be made to flow in a circuit when the circuit was moved 
relative to a magnetic field. According to the view held by Fara- 
day all these effects will involve electrical and magnetic media, 
and the idea grew that both electrical and magnetic effects might 
be considered to be due to the action of one medium, a kind of an 
electromagnetic ether. 

Now to set up strains in a medium in space requires that the 
medium should have elastic properties, just the same kind of prop- 
erties as the ether suggested for light. Faraday was brought to the 
point of considering that one ether might suffice for all these pur- 
poses and suggested the hypothesis that light itself was an electro- 
magnetic phenomenon. 


Proof of Relation between Light, Electricity and Magnetism 

Under the dominance of this idea, Faraday began to devise 
experiments which would establish the truth of this supposition. 
He searched for this evidence for many years and was rewarded 
for his endurance in the end by the discovery of what is known 
as the Faraday effect (1845). A beam of polarized light passing 
through some solids or liquids is affected by a magnetic field in 
such a way that the plane of polarization of the beam is rotated 
by an amount depending on the intensity of the magnetic field. 
Faraday's search for the discovery of a similar influence of an 
electrostatic field upon light failed, but another British physicist, 
Kerr, in 1875, discovered that a solid or liquid becomes double- 
refracting, like Iceland spar, under the influence of an electrostatic 
field. This is known as the Kerr effect; the Kerr cell, based upon 
it, has been an important element in early television systems. 

Maxwell's Electromagnetic Theory of Light 

Faraday's ideas found their most concise expression in mathe- 
matical terms in Clerk Maxwell's electromagnetic theory of 
light (1865). It achieves the fusion of the optical, electrical and 
magnetic theories, presented earlier, into one new theory that 
comprises all these phenomena. Faraday's ether, the carrier of 
electromagnetic energy and Fresnel's ether, the carrier of light, 
become one and the same; but the new ether is no more an elastic 
solid ether, but more ethereal in nature. Maxwell maintains that 
light is propagated as a transverse wave in the ether. The ex- 
istence of a wave demands that some physical quantity varies 
periodically along the direction of propagation of the energy 
transferred. This quantity, which formerly was the density of a 
material ether, now is the electric and magnetic field in free space. 
To speak of light as a transverse electromagnetic wave, then, 
means that along the beam electric and magnetic fields exist which 
are directed at right angles to the direction of propagation and 
which periodically rise and fall in intensity from zero to a maxi- 
mum to zero and a minimum and to zero again. Since Faraday's 
electric and magnetic fields are made the variable quantities of 


the light wave, Maxwell's theory demands that electric and mag- 
netic disturbances of the ether will show the same properties as 
light and will be propagated with the same velocity, i.e., with 
the speed of light. 

This theoretical conclusion was verified experimentally by 
Heinrich Hertz in 1887. He found that electromagnetic waves 
were produced by a spark gap and could be reflected, refracted, 
diffracted, and focussed in the same manner as light waves and 

Table 4. Classification of Electromagnetic Waves and Values 
of the Wave Lengths. 

Designation Limits of Wave Lengths 

of Wave From To 

Long radio 10000 m - 1000 m 

Broadcast band 1000 m - 100 m 

Short 100 m - 10 m 

Ultra-short 10 m - 0.5 m 

Micro-waves 50 cm- 0.1 mm 

Infrared or heat 0.1 mm- 0.001 mm 

Visible light 8 X 10- 5 cm -4 X 10" 5 cm 

Ultraviolet light 4 X 10~ 5 cm- ca 10 6 cm 

X-rays " ca 10" 6 cm- ca 10 9 cm 

Gamma rays (radium) ca 10~ 8 cm- ca 10- n cm 

Secondary Cosmic rays 10" 11 cm 
Wave length to be associated with 

electron beams From 10~ 8 cm up 

that they were indeed propagated with the same velocity. These 
epoch-making experiments thus confirmed Maxwell's electromag- 
netic theory in the most beautiful manner and everybody was 
very happy indeed. ■ The mystery of light seemed to have been 
solved once and for all. 

It is quite beyond the scope of this book to trace the develop- 
ment of the study of electromagnetic waves since Maxwell's time, 
but the above table shows the extent of its ramifications. We 
have recorded here the whole gamut of electromagnetic wave 
lengths, almost all of which have been measured experimentally 
by some satisfactory procedure. 

: 98M""""" 

(Above) Influenza Virus, X 42,000. 

(Below) Type A, Influenza Virus, Shadow Technique, X 95,000. 

Courtesy James Hillier, RCA 
Courtesy University of Michigan 

Chapter 8 

The Discovery of the Electron 

In spite of the success of the electromagnetic theory, there 
were minor uncertainties to be cleared up. It was not clear, for 
instance, how dispersion of light in a refracting body came about. 
In order to develop a satisfactory explanation for these phe- 
nomena, L. Lorenz (1880) postulated, in all material bodies, the 
existence of small, electrically charged particles, which were sup- 
posed to have the property of vibrating with a natural frequency 
when disturbed by a light wave. The creation of light at the 
source, i.e., the excitation of the ether, also was ascribed to such 
vibrating charged particles. They were, as yet, of an entirely 
fictitious nature from an experimental point of view, but their 
existence was demanded by a brilliantly conceived and well- 
founded theory. 

It was not long before the physical reality of these charged 
particles was established. Early in the second half of the nine- 
teenth century several outstanding physicists in various countries 
devoted a good deal of their time to the study of gaseous dis- 
charges, i.e., the passage of electricity through a partially evacu- 
ated glass envelope. Tubes known as Geissler tubes on the con- 
tinent and Crookes tubes in England were the earliest examples of 
these experiments, and our neon signs are an end product of this 
chain. During this period we meet men like Johann Wilhelm 
Hittorf, Sir William Crookes, Eugene Goldstein, Sir J. J. Thom- 
son, and F. Braun. Their work led to the discovery of the elec- 
tron by Sir J. J. Thomson in 1897. All the workers which we 
have just mentioned had found that a certain radiation is given 
off by the cathode, the negative pole, when a voltage is applied 
through the partially evacuated tube by connecting its sealed-m 




terminal wires to a Wimshurst machine, a battery of galvanic cells 
or an induction coil. This radiation came to be known as the 
cathode stream or cathode rays. 

Fig. 43a shows the diagram of a Geissler tube and the striations 
visible at a pressure of about 7.6 mm. of mercury.* Fig 43b 




— v Anode 

Simple gas discharge tube. 


The "Braun" discharge Tube. 



Fig. 43. 

shows the familiar Braun tube, a sample of which is in almost every 
physics laboratory. It is generally about one and a half feet long. 
At K an aluminum disc is sealed into the neck of the tube to serve 
as cathode and at A a wire is sealed into the side of the neck to 
serve as anode. A disc with aperture at B permits passage of the 
central section of the cathode stream which emanates from the 
cathode, K. This fine beam then strikes a mica disc, S, which is 
covered with some luminscent powder that gives off light when 
struck by the cathode stream. 

The Braun tube contains the essential elements which are 
present in a modern cathode ray tube, i.e., cathode, anode, and 
fluorescent screen. If the disc, B, were made the anode and the 
terminal at A left out altogether the similarity to modern tubes 
would be more evident. It is only because Braun tubes still con- 

* Low pressure in evacuated tubes is expressed in terms of "Millimeters Mercury" 
(mm. Hg.). The normal pressure of the atmosphere at sea level is equal to 760 mm. 
Hg. 7.6 mm. Hg. is then equal to T fo atmosphere. Modern radio tubes are exhausted 
to about too^otjo mm. Hg. or less. 


tain a lot of air that the position and shape of the anode does 
not matter. The very simple type of cathode shown in Fig. 43b 
is operative for the same reason. These gas discharge tubes lent 
themselves to a variety of interesting experiments. Sir William 
Crookes, E. Goldstein and others were untiring in their efforts to 
glean the true nature of the emanation from the cathode which 
Goldstein first called by the now familiar name of cathode rays. 
These scientists produced hundreds of tubes, many of which have 
been preserved to our day. By changing the shape of the elec- 
trodes and that of the envelopes, by varying the pressure of the 
remaining gas content and by other modifications, they unfolded 
a story which challenged the imagination of many researchers of 
their time. 

The Properties of the Cathode Stream 

There were many ways in which the cathode stream re- 
sembled ordinary light. When the stream was allowed to fall on 
the surface of various minerals placed inside the vacuum tube, the 
minerals would be made to fluoresce, often with brilliant colors. 
Sir William Crookes observed the phenomenon on such minerals 
as zincblende, fluorspar, willemite and diamond. When the high- 
est vacuum was reached, the glass of the tube itself gave a brilliant 
fluorescence wherever the cathode stream struck it; the color 
was usually of a greenish hue, but the color was found to depend 
on the nature of the glass. 

Fig. 44. 

The shadow of a metal 
obstacle cast by cathode 

If an opaque object were placed in the path of the stream a 
very sharp shadow was cast (Fig. 44). This experiment suggests 


that the cathode stream travels in straight lines as in the case of 
ordinary light. 

Goldstein found that when a coin was used as the metal 
cathode, instead of a plain smooth disc of metal, the imprint of 
the embossed figure on the coin, heads or tails, became visible on 
the fluorescent screen. This phenomenon suggests some kind of 
focussing effect on the screen due to the structure of the surface 
of the cathode. 

But there were two very important respects in which the 
cathode stream differed definitely from light. The stream was 
repelled from a negatively charged plate and attracted toward 
a positively charged plate — which suggested that in some way 
the stream bore a negative charge. Again, if an ordinary small 
bar magnet were brought near the discharge, the stream was 
deflected at right angles to the direction in which the magnet was 
pointed. If the north pole of the magnet were brought near the 
stream, the deflection would be in one direction; for the south 
pole, the deflection would be in the opposite direction. These 
experiments are easily shown with the Braun tube (Fig. 43b) as 
the motion of the beam is indicated distinctly by the movement 
of the fluorescent spot on the screen. No experiment was known 
which would show that ordinary light would be acted upon by 
either an electrostatic field or a magnetic field in such a simple 

The last experiment was epoch-making, because from it Sir 
J. J. Thomson deduced that the deflection was such as would 
happen if the cathode stream really consisted of a stream of nega- 
tively charged particles; these were afterwards called electrons. 

In addition to the properties already ascribed to the cathode 
stream, such as producing fluorescence, casting shadows and 
therefore travelling in straight lines, there are other properties of 
an electron stream which are of great interest. 

We believe that we do not see the electrons or the electron 
stream; what was seen and was given the name of cathode stream 
was a visible path which the electrons produce in knocking against 
air molecules. The end of the electron beam is indicated by the 
fluorescence on a luminescent screen placed in the tube. It is by 


the use of these screens that actual measurements of the motion 
of electron beams can be made. 

When the metal cathode was formed in the shape of a concave 
cup, the cathode stream was brought to a focus in the tube; if a 
piece of platinum foil is placed at this focus it soon becomes red 
hot — a fact first demonstrated by Crookes. The electrons give 
up their kinetic energy to produce heat. 

When electrons impinge upon a photographic plate they set 
free the silver from the silver bromide film in the same way as 
light does. The amount of blackening thus produced by elec- 
trons on a film indicates the intensity of the beam. This effect is 
utilized in high-energy cathode ray tubes in order to obtain a 
permanent record of very fast-moving beams. The pictures pro- 
duced by the electron microscope are obtained in the same 

The Charge and Mass of an Electron 

If one conceives that electrons are material, charged particles 
the question at once arises, "What is their mass and their charge?" 
Among the early experiments on the nature of the cathode stream, 
one offered an almost insuperable difficulty to the "material 
particle", theory. 

Lenard removed the fluorescent screen from the Braun tube 
and bored a very small hole in the large end of the tube where 
the cathode stream could be made to strike the tube. Over this 
hole he cemented a very thin metal foil window; this window 
sufficed to prevent gas molecules, even the smallest molecules, 
hydrogen, from entering the tube. In other words the vacuum 
could readily be maintained. Lenard got the curious result that 
the cathode stream came out through this window and gave a 
light spot on a fluorescent screen held outside the tube. Ordinary 
light can go through such thin foil; but if the cathode stream 
consisted of particles, these particles would have to be much 
smaller than the smallest gas molecule. To find the mass of the 
electron was then of prime importance. 

J. J. Thomson faced this dilemma and set about finding def- 
inite values for the mass, usually denoted simply by m and the 



charge, written e. The scope of this book does not allow us to 
develop this work in detail: suffice it to say that his experiments 
made use of the deflection of electrons in electric and magnetic 
fields. We shall refer only to the deflection in an electric field — 
that caused by inserting two metal plates in the Braun tube and 
providing means of keeping them at a difference of potential 
(Fig. 45b). 

The manner in which m and e came to be related in the ex- 
perimental results can be illustrated by referring to a simple 
problem in ordinary elementary physics. 

Fig. 45a. Gravity tends to pull the stone downwards during its flight. 

Suppose a boy standing on the edge of a steep bank is throwing 
stones at a sign board across a creek and aiming at a bird at A 
which is on his own level (Fig. 45a) . He soon finds by experi- 
ment that if he is to hit A he must give an upward direction to the 
stone at first. If the stone is thrown horizontally it will always 
strike at a lower level than A, such as the point B. The physicist 
expresses the relation between the mass, m, of the stone and the 
force of gravity acting on the stone by a simple statement that the 
force is given by the product of the mass and the acceleration, 
a, (the rate of gain of the velocity in the vertical direction). 
In our mathematical shorthand, we express this: f^mXa. The 
value of the acceleration can be easily determined if we know the 
time it takes the stone to travel from P to B and the distance AB. 



In the Braun tube (Fig. 45b) we suppose that the electron is 
shot in a horizontal direction from F. For the part of its journey 
that lies between the charged plates, the electron is exposed to 
the action of the electrical force acting between the charged 
plates, as a consequence of which it travels over the path, P'B'. 
The speed of the electron under the circumstances outlined is so 
great that the action of gravity on the electron may be neglected; 
in the arrangement indicated in the figures, the electrical force in 
Fig. 45b deflects the electron from a straight-line course just as 
the force of gravity in Fig. 45a deflects the stone from a straight 
line. We have, then, an electron of mass, m, acted on by a force, 


Fig. 45b. Deflection of electron beam in an electric field. The part of the path 
between the plates is curved. 

say f, and given thereby an acceleration, say a. We may then 
write: f = mXa. 

In the case of the electron the force, f , depends directly on the 
size of the charge, e, and the expression for the acceleration in- 
volves the velocity of the electron and the distance A'B'. As we 
shall see later, the force, f, on the charge, e, is given by the rela- 
tion: f=FXe, where F is the strength of the electrical field acting 
between the plates, which in turn is very simply determined in 
the experimental arrangement here used. So in the relation, 
f = mXa, we may substitute (FXe) for f which gives the expres- 
sion mXa = FXe. By a simple transformation we may write this 
in the form e a 

m F 


Since both a and F can be measured experimentally, we thus 
arrive at an experimental value of the ratio e:m. 

The foregoing indicates roughly the manner in which J. J. 
Thomson approached the problem. The upshot was that these 
early experiments gave a definite value to the ratio of the quanti- 
ties e and m, but did not settle the definite value of either of them 
separately. Historically the next step in the development was to 
assume a definite value for one of them. This choice fell on the 
the value for the charge e. 

Faraday's early experiments on the conduction of electricity 
by liquids, known as electrolysis, seemed to indicate that there 
existed a small ultimate unit of electrical charge — an atom of elec- 
tricity, one might say. The magnitude of this unit was equal to 
the charge on a hydrogen ion (which is positively charged) or to 
the charge on a chlorine ion (which bears an equal negative 
charge) ; the numerical value of this ionic charge had been satis- 
factorily determined long before 1897. Using this value for e, 
the value for the mass of the electron came out to be about 
l/2000th of the mass of an atom of hydrogen. It required a great 
flight of the imagination and a considerable amount of intellectual 
bravery on the part of J. J. Thomson to announce his belief in 
the existence of a material particle of such unheard-of minuteness. 
In the years that followed everyone came to regard this strange 
conclusion as a fact. 

The value of the charge on an electron is now accepted to be 
4.804 XlO 10 electrostatic unit, or 1.603 XlO' 19 coulombs. This 
value is, of course, also the value of the charge on a univalent 
electrolytic ion, such as that of hydrogen or chlorine. 

The deduced value of the mass, m, of the electron then turns 
out to be 9.1 XlO" 28 gram. The mass of the hydrogen atom is 
1.675 X 10~ 24 gram, from which we see that the ratio of these two 
masses is approximately 1:1838. 

The Electron as a Fundamental Particle 

It is to be noted that electrons are produced in many different 
ways, and experiments similar to J. J. Thomson's were carried 
out to determine the value of the ratio e/m for electrons produced 


in a great variety of ways. The amazing result of all these in- 
vestigations, carried out by many workers in different countries, 
was that, no matter how the electrons were produced, they always 
gave the same value of the ratio e/m. Moreover, although 
Braun tubes were made with a number of different metals for 
the cathode, all these metals gave off electrons having the same 
e/m ratio. It is true that some metals will give off their electrons 
more readily than others, but there is no doubt that all metals 
contain a copious supply of electrons. 

Later experiments have disclosed that insulators also contain 
electrons, but they are not nearly as readily released from the 
confines of these materials, be they glass, ebonite, sulfur or any 
other insulating substance, as in the case of metals. 

We may then look upon the electron as a building stone of 
matter in general. Even organic materials, such as plastics, wood, 
cotton, and plant tissue contain electrons in their basic atomic 

The Building Stones of Matter 

It would be misleading to say that all matter is made of elec- 
trons only. It is true that all matter contains electrons, but they 
are not the only fundamental building stones. Since the value of 
e adopted for the electronic charge was equal in magnitude to 
the charge on a hydrogen ion, and so related to the hydrogen 
atom, it was thus postulated that the hydrogen atom was tied up 
in some way with one electron. Now it was known already 
from earlier experiments that atoms (or even molecules) are 
not, of themselves, charged either positively or negatively. Con- 
sequently, if we assume one electron to be a constituent of a 
hydrogen atom, we must assume that there will be another part 
of the hydrogen atom which possesses, of itself, a positive charge 
equal in magnitude to e. This assumed second part of a hydro- 
gen atom has been given the name of proton. 

Since a hydrogen atom consists of one electron and one pro- 
ton, and since the mass of one electron contributes only about 

th of the total mass of the hydrogen atom, we are led to say 

2000 J 5 


that the mass of the proton is about n ths of the mass of the 

hydrogen atom. 

Early in the last century there was propounded the theory 
that all other atoms were made up of hydrogen atoms. In place 
of this we now say that atoms — all atoms — are made up funda- 
mentally of electrons and protons, in equal numbers. The elec- 
trons contribute little to the mass. Elemental substances differ 
from one another only in the number of electrons and protons 
per atom. 

There is very definite experimental evidence that when strong 
electrical forces from external sources are exerted on a group of 
atoms, some of the atoms are broken up, giving rise to positively 
charged and negatively charged remnants. Since the electron 
has such a small mass, the idea was suggested that this breakup 
of atoms was due to the expulsion of one or more electrons from 
the atom. Some atoms thus lose one or more of their units of 
negative charge and are then left with a net positive charge equal 
in magnitude to that of the negative charge carried by the escap- 
ing electrons. An atom thus deprived of one or more of the 
normal number of its electrons is called an ion, and the process 
is known as ionization. 

This illustrates the beginning of the construction of a model 
of an atom. It should be apparent to all that the whole develop- 
ment is built up on step-by-step assumptions suggested by our 
everyday experience with things which we handle. 

Electron Emission in the Braun Tube 

We are now in a position to say a few words about the origin 
of the electrons in the Braun tube and the mechanism by which 
they are produced. The pressure in this tube, as first constructed 
by F. Braun in 1897, is about 5 millionths of one atmosphere or, 
in terms of millimeters of mercury, between 1/100 and 1/1000 
mm. At this pressure one cubic centimeter contains millions of 
molecules made up of some of each of the constituents of air, 
namely oxygen, nitrogen, carbon dioxide, hydrogen, carbon 


monoxide and the rare gases argon, neon, krypton, xenon and 
radon to a very small degree. The molecules of most of the 
gases consist of two or more atoms that normally associate in 
groups and they are almost always electrically neutral. 

Ordinarily some of these gas molecules are ionized, since 
cosmic rays and rays from radioactive materials traverse all space 
and even penetrate the glass of the tube. These rays have a great 
content of energy which they can impart to the gas molecules 
with the result that, in the space between the cathode and the 
anode, some electrons are set free and positively charged gas ions 
are created. The electrons will move toward the anode and the 
positive ions toward the cathode. It is a peculiarity of gas-dis- 
charge tubes that the strongest electric force exists in the region 
close to the cathode. The fairly heavy positively charged ions 
drift toward the cathode, due to the electrostatic attraction, and 
they are finally given a great acceleration toward the cathode disk 
and impinge on it with great energy. This energy sets free elec- 
trons from the metal of the cathode and these electrons in turn are 
now driven away from the cathode by the electrostatic repulsion. 
They shoot like bullets out of a gun and travel in approximately 
straight lines toward the fluorescent screen. If such an electron 
should strike a gas molecule during its travel it will give up some 
of its energy and ionize this molecule, creating new electrons 
and ions, so that in the end a regular avalanche of positive ions 
strikes the cathode, producing more electrons. The bombard- 
ment of the cathode by positive ions can be so severe that some 
of the metal is sputtered away and deposits as a thin film on the 
glass wall of the tube. 

It is readily appreciated from this somewhat sketchy descrip- 
tion that the Braun tube operates by virtue of its imperfect 
vacuum. If there were no gas left in the tube the avalanche 
could not get started and no electrons would be obtained. On 
the other hand, we have all complained about gassy radio tubes 
in our sets at one time or another and so have an appreciation of 
the undesirability of gas-filled tubes. There must be other ways 
to get electrons out of cathodes. Let us find out how. 



Courtesy University of Toronto 

Lens paper and methylene blue. A solution of methylene blue was blown through an 
atomizer into air and the air was then drawn through lens paper. The bare parts are holes 
in the paper. The bar across the middle is a very fine fibre of the paper. Note the spherical 
shape of the methylene blue particles and the strange way in which they cling to one another. 
X 14,200. 

Chapter 9 

Thermionic Emission from Metals 

It has been mentioned above that all matter contains electrons. 
Among the solid substances, metals in particular show an abun- 
dance of electrons, and it is because of this fact that metals "con- 
duct electricity." This latter statement simply implies that some 
of the electrons contained in the metal are free to move under the 
influence of an electric field established within the metal by the 
application of an electric force. The flow of electrons within 
metallic conductors drives our motors, lights our lamps and heats 
our kitchen ranges. 

The flow of electrons through empty space in a vacuum tube 
actuates radio tubes, photocells, cathode ray tubes, television 
camera tubes and electron microscopes. Evidently in these cases, 
some means must be found always to get the electrons out of the 
metal and make them travel through the vacuum along predeter- 
mined paths. 

The most commonly employed method by which electron 
emission is obtained from a metal is that of heating the metal to 
a high temperature. That electrons are emitted from an incan- 
descent wire in the presence of an external electrical field was 
discovered by Thomas Edison in 1881, though not described in 
our present-day language. Broadcasting transmitting tubes utilize 
incandescent tungsten filaments as a source of electrons. Wehnelt 
discovered in 1904 that the oxides of such metals as calcium, 
strontium and barium emit electrons more copiously at much 
lower temperatures than metals do. These "dull emitters," oper- 
ating at a dull red heat, are used nowadays in all our radio receiv- 
ing tubes. They consume less power than a bright emitter and 



consequently are more economical in operation. Dull emitters 
yield more electrons per dollar. 

Electron emission obtained from metals by heat is described 
in technical language by the term thermionic emission. This 
name was coined at a time when the nature of the emitted particles 
was still in doubt and it does not describe thermal electron emis- 
sion clearly. To be sure, there always takes place a definite, 
though negligible, amount of ion emission when electrons are 
emitted, but the term thermionic emission is now used with the 
meaning of thermal electron emission, i.e., emission of electrons 
by heat. 

The emission of electrons from a metal surface has, at times, 
been compared to the evaporation of water vapor or steam from a 
water surface. At reasonably low temperatures, the water par- 
ticles cannot escape very easily from the bulk of the liquid since 
they are held in subjection by the attraction of their fellows. 
When the water is heated to higher and higher temperatures, 
some individual water molecules gain enough kinetic energy to 
escape from the surface. When the temperature has risen to the 
boiling point, all the energy put into the water as heat is used to 
provide escape energy for the molecules at the water surface 
until all water has evaporated. We picture this process as a 
mechanical model for thermionic emission. In this case also heat 
must be provided in order to overcome forces at the surface of 
the metal which tend to restrain the electrons. 

Electron Emission from Oxide-coated Cathodes 

Thermionic emission from oxide-coated cathodes has been 
utilized technically for over thirty-five years, but a clear under- 
standing of the mechanism of thermionic emission has been gained 
only recently. It has been established by innumerable experi- 
ments and investigations during the past few decades that the 
copious emission of electrons from an oxide-coated cathode is 
due to the existence of a very thin film of metal on the surface of 
the oxide layer. 

The oxide layer itself is produced by a chemical reaction 
that takes place when the cathode is first heated in vacuo. It 


is probably the most general practice in America to spray onto 
the cathode surface a mixture of barium carbonate (BaCO s ) and 
strontium carbonate (SrC0 3 ), to which calcium carbonate 
(CaCO s ) is added in some cases. These carbonates are prepared 
in the form of a suspension in amylacetate, with some nitro- 
glycerin as binder. Mixtures of this type are commercially avail- 
able from chemical supply houses. The cathode material consists 
of pure nickel or nickel with a fraction of a per cent of titanium, 
aluminum or magnesium added. During the process of exhaust- 
ing the tube the cathode is heated in steps up to a temperature of 
about 1350° K. # The alkaline-earth carbonates (or nitrates, or 
hydroxides as the case may be, depending on the compound that 
was originally sprayed onto the cathode) break down at about 
1100° K into oxides and carbon dioxide (or nitrous oxide or 
hyrogen in the case of the nitrates and hydroxides, respectively) . 
The gas is pumped away, leaving the oxide layer on the cathode. 
In the presence of small amounts of reducing agents such as car- 
bon, titanium, aluminum or magnesium, about 0.5 per cent of the 
oxide molecules are reduced to free barium (or strontium or 
calcium), with the release of oxygen. Confining our description 
to the case of barium, we may say that the barium atoms in the 
bulk of the oxide coating find their way to its surface by diffusion 
and form a monatomic layer which, for optimum emission, need 
not cover the oxide surface completely. Partial reduction of the 
oxide takes place at the "sinter temperature" of about 1350° K. 
During its use the cathode is operated at about 1000° K and dif- 
fusion of the atoms to the surface may continue during its life. 
This thermal activation in the absence of externally applied volt- 
ages, which we have just described, is possible only in the presence 
of reducing agents. It is a reliable technique for mass production 
and does not require any prolonged ageing of the cathode after 
the tube has been exhausted although ageing is generally applied 
in order to stabilize the emission. 

* 1350° K is read "1350 degrees Kelvin" or "1350 degrees absolute" and refers to 
the absolute temperature scale introduced by Lord Kelvin. This scale is widely used 
in the scientific literature and can be translated into centigrade by subtracting 273.18 
degrees; thus 1350° K equals 1076.82° C. Or, the other way around, °K is obtained 
from °C by adding 273.18. Thus, water boils at 373.18° K under normal pressure. 


Aside from this purely thermal activation, another activation 
procedure may be followed which is characterized by the appli- 
cation of anode voltage during the high-temperature treatment of 
the cathode while the tube is being evacuated. It may then be 
assumed that the oxide is separated into barium ions and oxygen 
ions which will migrate under the influence of the applied field 
according to Faraday's laws of electrolysis. The direction of 
the field is such that the positive barium ions drift through the 
oxide toward the core metal, where they give up their charge, 
while the negative oxygen ions migrate toward the surface of the 
coating and escape. The neutral barium atoms will then again 
diffuse through the oxide coating from the core metal to the 
surface and form the active surface layer, which is the origin of 
the electron emission under normal operating conditions. This 
electrolysis activation does not require any reducing agents and 
can thus be applied to materials of extreme purity. 

The existence of the thin barium film, the way in which 
it is produced, and the seat of the electron emission, have been 
subjects of lively controversy for many years. However, very 
conclusive experiments have been carried out that prove the exist- 
ence of the barium film at the cathode surface and show that 
without its presence the copious electron emission ceases; the 
barium film evaporates at 1600° K. The electron microscope 
has added greatly to a better understanding of thermionic emis- 
sion, since it not only permits visual observation of the intensity 
distribution of the electron emission over the cathode surface 
during operation, but also reveals minute changes in emission 
from point to point, at the surface of the coating. 

Comparison of the Electron Yields of Emitters 

Pure metals, such as nickel, tungsten, molybdenum and plat- 
inum, do not emit electrons unless they are heated to very high 
temperatures. Only tungsten and molybdenum are used as 
"bright emitters" in practical cases. To give an example, an 
oxide-coated cathode will emit 100 ma. per square centimenter of 
its surface when at a temperature of 1000° K. It will require a 
certain number of watts to reach and maintain this temperature. 


If we divide the total emission obtained from a cathode by the 
number of watts supplied to the cathode heater, we arrive at the 
emission yield, which is about 20 milliamperes per watt for an 
oxide-coated cathode. A tungsten filament operating at 2300° K 
will give about the same emission per sq. cm as the oxide cathode 
at 1000° K, i.e., 100 milliamperes per sq. cm, but the emission 
yield will be less than one milliampere per watt. 

In order to produce an electron current between cathode and 
anode it is necessary to apply a positive voltage to the anode. 
We shall explain very shortly (p. 135) that an accelerating field 
must exist at the cathode surface in order to send the electrons 
on their way toward the anode after they have been released 
from the bulk of the cathode by the thermal energy imparted to 
them from the heater unit. 

The ease with which a cathode surface is able to emit elec- 
trons is expressed by a term called "work function," the symbol 
for which is 4>. It stands for the amount of work which the 
electrons have to perform in escaping from the cathode against 
the restricting forces exerted by the atomic lattice of the cathode 
and the induced electrical forces immediately after their escape. 

The lower the value of ^>, the more copious will be the emis- 
sion under any given conditions, and the lower will be the oper- 
ating temperature of the cathode necessary to obtain a certain 
required emission current. Table 5 gives the value of the work 
function, (/>, in electron volts for a number of different cathode 
materials and also the common operating temperature in absolute 
degrees (°K), the melting point in °K, the emission yield in 
milliamperes per watt and the electron emissivity in milliamperes 
per sq cm of active surface. 

Secondary Electron Emission 

We have so far described two mechanisms by which electrons 
may be released from the confines of a metal. In each case it 
was necessary to add energy to the cathode. The "cold" cathode 
of a gas-discharge tube was bombarded by positive ions which 
imparted their kinetic energy to the cathode; the "hot" cathode 
of Wehnelt received energy in the form of heat. In both cases 



this imparted energy was transformed into kinetic energy gained 
by the "free" or "conduction" electrons of the cathode, which 
were thus enabled to "escape" from the cathode into the vacuum. 
This "escape energy" can be provided in still different ways, 
as we shall see presently. 

We may well utilize again the model introduced above which 
established an analogy between the evaporation of water mole- 
cules and thermionic emission of electrons. Small masses of 
water can be forcibly separated from any large body of water, 
even at low temperatures, when a foreign body, for instance a 
stone, is thrown into the water and a splash is produced. The 
splashing water spray derives its escape energy from the kinetic 

Table 5 












ma/cm 2 













Barium oxide 







Strontium oxide 






Calcium oxide 



























Thoriated tung- 







energy of the stone that has been thrown in. When particles 
much larger than electrons, such as positive ions, are directed 
toward a metal surface with sufficient velocity, their impact will 
"splash" electrons from the metal. Likewise, if electrons of 
sufficient velocity, obtained from another source, are directed 
toward a metal surface they will impart their energy to the free 
electrons of the metal and thus knock them out of the metal. 
Electrons emitted by the bombardment of a surface with charged 
or uncharged particles are called secondary electrons, and this 


type of emission is termed secondary emission. The bombarding 
particles, if they are electrons, are called primary electrons. 

If we carry our analogy a little farther we may gain greater 
insight into the mechanism of secondary emission. The splash of 
water is noticeably more pronounced when the stone is thrown 
into the water at an oblique angle than when it is simply dropped 
vertically. Similarly, the secondary electron yield is greater 
when a primary electron strikes a metal surface at an oblique 
angle rather than at a right angle. Two, three, indeed as many 
as eight secondary electrons may leave the surface for each 
impinging primary electron; and finally, as the stone may skip 
after its first contact with the water, the primary electron itself 
may be "reflected" by the metal after a very slight penetration 
into it and be re-emitted as a reflected secondary electron. 

All these effects have a great practical importance. During 
the early stages of the development of electronics, the science 
dealing with electronic tubes and devices, secondary electrons 
had none but nuisance value. They were most unwelcome, 
since they interfered with the controlled electron stream emitted 
by the cathode; they had to be suppressed and the introduction 
of the "suppressor grid" into the radio tube which promoted the 
triode to the tetrode was a distinct advance. In later years, ways 
and means were found, however, to utilize secondary electron 
emission to advantage so that today secondary emission amplifiers 
are available which have made possible great advances in tele- 
vision. One may well say that without secondary electron 
emission present-day cathode ray television systems would not 
be possible. 

Photoelectric Emission 

Photoelectric emission of electrons is another example of cold- 
cathode emission. Electrons are released from a metal surface 
by the incidence of light. This effect was discovered by Wilhelm 
Hallwachs in 1888. Since we used the splashing of water for 
the explanation of cold-cathode emission by bombardment 
with positive ions or electrons, we may well extend it to photo- 
electric emission. The stone or foreign body which produced 


the splash of droplets when dropped or thrown into the water 
was the model for positive ions in one case and for primary 
electrons in the other. We shall now let it stand for Newton's 
light particles and thus find no difficulty in forming a picture of 
the photoelectric effect. In all three cases we transfer the kinetic 
energy of a particle to the free electrons of the metal and thus 
provide the necessary escape energy for their release. 

By reinstating Newton's light particles we apparently nullify 
our earlier statement that these could not stand the test of experi- 
mental facts discovered later. It is quite true that the particle 
concept did not fit the effects of diffraction, interference and 
polarization, all of which were discovered long before 1888. 
The photoelectric effect, on the other hand, presented an entirely 
new problem, as we shall show presently. It can be explained 
satisfactorily only on the basis of the particle concept, and the 
extension of our analogy above is therefore quite justified. 

Luminescence and X-rays 

We have just learned that a beam of light can release electrons 
from a solid body. Sir William Crookes demonstrated unknow- 
ingly that a beam of electrons can release light from certain 
minerals when these are bombarded by electrons. The lumines- 
cent screen of willemite in the Braun tube is based on this effect. 
Very fast electrons will release an invisible light from metals 
when they impinge upon them. This was discovered by Roent- 
gen in 1895, and the invisible radiation emitted by the metal is 
now known under the name of x-rays. It was established by 
experiments in later years that x-rays have the same fundamental 
properties as ordinary light; their energy content is, however, 
very much greater. They are thus not so readily absorbed by 
any material bodies that stand in their path. They penetrate the 
glass wall of the tube envelope within which they are produced 
and are stopped only by fairly dense substances, such as bone 
structures or metals. When the x-rays impinge upon a photo- 
graphic plate they reduce the silver bromide film in the same way 
as light does and thus produce a shadowgram of the objects in 
their path. 



Micronex Carbon 
Black Particles. 

RCA Laboratories 


P-33 Carbon Black 
Particles; note smooth 
surface and spherical 

Courtesy Columbian Carbon Co. 



mmWf : - 




-;''•, '¥B#fe#' 



. ■ \ .. 


Courtesy Stanford University 

"Extra Light" Magnesia, 70 kv. original magnification X8200. 

Chapter 10 

The Photoelectric Effect 

During his experiments on the creation of short radio waves 
from an electric spark, Heinrich Hertz observed in 1887 that a 
spark would pass the gap more easily if ultraviolet light fell on 
the negative pole of the gap. Stimulated by this observation, 
Wilhelm Hallwachs, a year later, found that a negatively charged 
body loses its charge under the influence of ultraviolet light; 
this is the so-called Hallwachs effect. In later years, it was 
shown that a great number of substances, solid, liquid and gaseous, 
emit electrons under the influence of light. This phenomenon 
is called the photoelectric effect. It is the basis of the modern 
"magic eye" or photocell. Television pickup cameras are based 
on the photoelectric effect and so are sound films. Innumerable 
industrial devices utilize photoelectric cells which are now being 
manufactured in large quantities. Aside from its practical impor- 
tance, the discovery of this effect marks a milestone in the history 
of physical science. It led to the establishment of the dual theory 
of light. Since this subject is our main concern in this chapter, 
we must of necessity take pains to study the laws that govern 
this photoelectric effect. 

Let us first become familiar with the physical appearance of a 
photocell and its operation. Fig. 46 shows the diagram of such 
a cell in its simplest form. A spherical glass envelope with a 
tubular neck, St, is mounted in a base, B, which carries a base pin, 
P. This pin is connected with a straight wire, A, (or a loop), 
that is supported in the center of the sphere on a glass stem. 
The inside of the glass sphere, with the exception of a window, 
W, is silvered and electrical connection with the silver film is 
made through a fine wire which is sealed into the glass and 




soldered to the terminal, K; this film serves as the cathode and 
A as the anode. During the process of manufacture the cell is 
connected to an exhaust system through a tube attached at T. 
The silver layer is oxidized by passing a glow discharge in an 
oxygen atmosphere and then cesium (or some other alkali metal) 
is deposited on the silver oxide. A carefully controlled baking- 
process permits a reduction of the silver oxide and the formation 
of cesium oxide with some excess of cesium at the inner surface. 
This complex surface, the photo-cathode, has been found to yield 
a large supply of electrons when it is exposed to a light source, L, 
sending its rays through the clear window, W. In order to facili- 
tate the flow of electrons through the cell, a positive voltage, E, 
from a battery is applied to the anode, A; a current, i, indicated 
by a micro-ammeter, M, will flow through the circuit as soon as 
the cell is exposed to light. When the light is shut off, or the 
window covered, the current falls to zero immediately. 






Fig. 46. A photoelectric cell. 



Fig. 47 gives the circuit diagram. A load resistance, R, is 
included in the circuit. This may take the form of a sensitive 
relay, which operates any desired device, or the voltage drop 
across the resistance, R, created by the photo current, i, may be 
fed to the input stage of an amplifier so that a less sensitive relay 
may be used in the output circuit of the amplifier tube. 

The following facts can now be observed: 

(1) The photo-electrons are emitted by the cathode, with- 
out any measurable delay in time, as soon as the light reaches it. 

(2) The number of electrons emitted increases with an 
increase in the intensity of the light source; i.e., the stronger the 
illumination of the cathode the greater the number of electrons 
emitted per unit area of the cathode and the greater will be the 
deflection of the ammeter. 

(3) The frequency of the light source, or, in other words, 
the color of the light or the wave length of the light, has a decisive 
effect on the operation of the cell. The velocity with which the 
electrons leave the cathode, or, in other words, the energy of the 
electrons at the moment they leave the cathode, increases with the 
frequency of the light. The intensity of the light has no effect 
on the value of this initial energy of the electrons. Furthermore, 

Fig. 47. A photoelectric cell circuit. 



the frequency of the light source must exceed a certain character- 
istic minimum value before any electrons are emitted at all. This 
threshold frequency, v , or threshold wave length, X , depends on 
the type of cathode that is used in the cell. It is thus possible to 
make cells that are particularly sensitive to red light, others that 
respond to blue or ultraviolet light, and again others that have 
their maximum sensitivity coincide with that of the human eye 
at 5,500 A. U. (yellowish green) . 

The laws governing the photoelectric effect are illustrated 
by graphs in Figs. 48, 49, and 50. In Fig. 48 the current, i e , car- 
ried by the number of electrons, n, emitted per second per unit 

Fig. 48. 

Relation between the 
photoelectric current, i e , 
and the intensity of the 
light causing this current. 

Light Intensity 

of cathode area is plotted as ordinate and the incident light inten- 
sity, i u as abscissa. We find that i e is directly proportional to i t . 
Fig. 49 shows how the energy, E, of the emitted electrons 
increases uniformly as the frequency, v, of the light from a con- 
stant-energy light source is gradually increased. It is to be noted 
that a certain definite theshold frequency, v , must be reached 
before any energy is emitted. Fig. 50 gives the response of a 
cell to the various wave lengths of the spectrum. The photo- 
current per unit of area which results from a unit of light energy 
of a given wave length is plotted for the whole spectrum of wave 
lengths. When the wave length is longer than X in the infrared 
region, corresponding to v in Fig. 49 (y = v/h) , no more electrons 
are emitted. This random example, Fig. 50, discloses a strong 
maximum photo-current in the blue region of the spectrum and 
a lesser second maximum in the red. 



The Inadequacy of the Wave Theory 

We shall now try to explain these facts on the basis of Max- 
well's electromagnetic wave theory of light. Without losing 
ourselves in too many details, we may apply the familiar concept 
of any wave theory to the problems at hand. It is, for instance, 
readily observed in everyday occurrence that a radiating source 
gives off energy uniformly in all directions. Thus, a light source 
can be seen from all directions, a bell on a church tower can 

Fig. 49. 

Dependence of the 
energy of the emitted 
electrons, E, and the fre- 
quency of the incident 
light. The frequency v 
is the point where the 

frcQuency of Light s ra P h leaves the axis - 

*- Wave Length of light 

Fig. 50. How the current, i, of a typical photoelectric cell varies as the wave 
length of the incident light increases. 

be heard from all directions, as long as no obstacle intervenes. 
The intensity of the light received or of the sound heard will 
decrease with the distance from the source according to a definite 
law. It turns out that the received energy falls to one-quarter 
of its original value when the distance is doubled or to one-ninth 
when the distance is trebled, and so on: in other words, the 
received energies are coupled with the distance from the source 
by the inverse square law. If we keep the distance from the 



source constant, the received energy at that distance will be the 
same per unit of time, no matter in which direction from the 
source the receiver is located. If an airplane flies in a circle 
around a transmitting tower, the light at its top will appear of 
a certain constant brightness to the pilot. He could fly on the 
surface of a sphere with the light source as center and still observe 
the same brightness; this sphere is indicated in Fig. 51. If the 
pilot then should fly further away, say, on a sphere twice as large, 
the beacon would appear to be only one-quarter as bright. If we 

Fig. 51. 

How energy received 
from a point source varies 
with the distance from 
the source. 

were to follow the light energy that is sent out by the beacon 
at a particular instant and trace its propagation through space, 
we would have to visualize something like a soap bubble that 
first snugly fits the light bulb and then expands continuously to 
enormous proportions. It would of necessity become thinner 
and thinner as its size increased; the thickness of the film may be 
taken to measure the energy contained in a unit area of the sur- 
face. Since the fixed amount of energy that started to peel itself 
off from the bulb is spread over a larger area when the size of 
the sphere has increased, the energy per unit area must of neces- 
sity decrease. 


When our light energy, or soap bubble — to continue the 
analogy — reaches a photo-cathode, the surface element of the 
bubble, say one square centimeter, will impinge on the corre- 
sponding surface element of the cathode and so transfer its energy 
to it. The electrons emitted from this small area by the light 
energy striking it should then share this energy among themselves 
and, according to the principle of the conservation of energy, the 
combined energy of the electrons could not exceed the energy 
content of the surface element of the light wave. Careful 
measurements disclose that this condition is not fulfilled. The 
electrons are actually emitted with a far greater energy than could 
be conveyed to them by a uniformly expanding spherical wave. 
It might be suggested that the cathode stores up the energy of 
a number of successive waves until the required amount is ob- 
tained to account for the observed electron energies. This 
would take considerable time, of the order of hundreds of hours, 
before the figures would agree. Actually, as has been stated 
above, no measurable time delay between the arrival of the 
light wave and the emission of the electrons can be observed. 

Our discussion has been based on two assumptions. First, 
the validity of the principle of the conservation of energy and 
second the wave nature of light. Since our deductions cannot 
explain the observed effects one of the assumptions must be 
wrong. The law of the conservation of energy has been found 
to be quite consistently true and serves as one of the pillars on 
which the science of natural phenomena seems to rest. We 
must then, once again, call into question the wave nature of light. 

The Quantum Theory 

The nineteenth century complacency regarding the suffi- 
ciency of the wave theory of light received its first shock in 1900 
through the work of Professor Max Planck of Berlin. He found 
it impossible to explain the experimental relations between the 
absorption and emission of radiant energy by hot bodies and their 
temperatures by any wave motion model. The classical concept, 
accepted at this time, was that the continuous flow of energy 
through space in the form of electromagnetic waves required that 


both absorption and emission of such energy by hot bodies 
should also be continuous. But this theory of absorption and 
emission could not be made to agree with experimental facts. So 
Planck made the exceedingly bold assumption that, no matter 
how the energy is transmitted, absorption and emission at mate- 
rial bodies takes place in very small but definite units, i.e., dis- 
continuously. Each small unit bundle of energy, Planck called 
a quantum; he suggested that there would exist an "atom of 
mechanical action," which came to be represented by the letter 
h, now known as Planck's constant. 

Now, when the curious results given by the photoelectric 
experiments were considered, it was apparent that they could be 
explained if one assumed a corpuscular theory for light. Einstein, 
in 1905, extended Planck's theory to say that not only are absorp- 
tion and emission discontinuous but even the energy of a light 
beam travels through space in quanta. This brings us back to 
the corpuscular theory of light — Newton's theory! 

A light corpuscle, a quantum or a photon, as it is called, would 
carry its energy undiminished from the source to the greatest 
distance and would be able to hand its energy to an electron when 
it strikes a photo-cathode so that the electron can be emitted 
instantly. The more intense the light source, the more corpuscles 
are emitted and the greater the number of electrons set free. 
This satisfies actual observations. 

In order to explain the relation between the electron energy 
and the frequency of the light in the photoelectric effect (Fig. 49) 
it becomes necessary to assume that the various corpuscles of light 
have different energy contents depending on the frequency of 
the light. Thus a corpuscle, or quantum, or photon, of blue 
light contains more energy than one of red light; the electrons 
set free by blue light move faster than those set free by red light. 

Einstein suggested that, as the energy is proportional to the 
frequency, it might be expressed by the relation: 

E(the energy) = (a constant) Xv (the frequency) 

He further suggested that the constant be put equal to Planck's 
constant. This leads us to the fundamental equation for a quan- 
tum: E = hv 


We now find ourselves in a great predicament. All the argu- 
ments raised against Newton's corpuscular theory of light must 
of necessity come up again. We know that a corpuscular theory 
cannot explain diffraction, interference, or polarization. We may 
recall that Newton himself hesitated to be tied down to the 
corpuscular theory exclusively and made use of periodic states 
which imply nothing else but a wave character. This intuitive 
hesitation of his to accept a one-sided theory is now raised to a 
basic principle in physics. Light is neither wave nor corpuscle: 
it is both. This is what we mean by a Dual Theory of Light. 

One should not be frightened at this seemingly contradictory 
statement. The fact is that we cannot bend nature to suit the 
niceties of our old-fashioned ideas. Things evidently are not as 
clear-cut as we would like them to be. We should furthermore 
bear in mind that the light corpuscle or photon, hv, is defined 
by a frequency, and thereby is closely tied in with the wave con- 
cept. Frequency is a derived term; it means the number of times 
that a wave recurs per second. To find the value of the fre- 
quency, we measure the wave length and then divide it into the 
velocity of light. We might say that the path of the photon is 
guided by the wave that is associated with it (Fig. 52) . It simply 
helps our mind to visualize what may be going on. A transverse 

Fig. 52. An electron and its associated wave. 

wave is drawn that weaves itself around the direction of propa- 
gation indicated by the axis. Wherever the wave intersects the 
axis the heavy dot represents a photon traveling in a straight line. 
In disclosing this dual nature of light, not by the choice of 
their fancy but driven by irrefutable facts, physicists now think of 
the wave concept when describing phenomena that are evidence 
of the wave nature of light, and of the photon concept when deal- 
ing with the photoelectric effect and radiation problems. In the 
words of Sir William Bragg: "We teach the wave theory on 


Mondays, Wednesdays and Fridays and the corpuscular theory 
on Tuesdays, Thursdays and Saturdays." 

When we review the changes that the concepts of the nature 
of light have undergone during the course of the years and realize 
the fruitfulness of the concept of duality, we may well turn to 
the electron and question whether it has an exclusively cor- 
puscular nature. It will be shown in the following chapter that 
for the electron also a dual theory has been firmly established. 
The wave nature of the electron is indeed the basis upon which 
the electron microscope rests. 


m • 

Courtesy Columbian Carbon Co. 

Particles of rubber in natural Particles in synthetic-rubber latex 

latex, magnified 25,000 times. (Buna-S) magnified 25,000 times. 




The first magnetic electron microscope constructed in America at the University of 
Toronto. At the right is shown part of the evacuating system and rheostats used 
to regulate the magnetic fields of the lenses. The overall height is about six feet. 
The first picture was taken in June, 1938. 














2. ■ "■ 
" ? " ;; »*Jc 


Courtesy R.C.A. Laboratories 

Aluminum Oxide (X 44,000), 

Chapter 11 

Waves and Matter: de Broglie' s Theory 

Although the dual theory of light was impressed on the 
physicist rather early in this century, the far-reaching conse- 
quences of the fundamental experimental facts on which it was 
based were not fully recognized until de Broglie of Paris 
announced his new views in 1923. We have seen how light must 
be looked upon as having both a wave nature and a corpuscular 
nature, in order that we may have a satisfactory model simulating 
the action found in the various light experiments. In the photo- 
electric effect we find the energy of light directly inducing 
motion in electrons. Some time after the discovery of this effect, 
A. H. Compton (1922) showed that certain experiments could 
be explained satisfactorily only by saying that the action of the 
radiation of light in the form of x-rays upon an isolated electron 
was the same as though there were an ordinary mechanical 
collision between two corpuscles; the two corpuscles Compton 
suggested were the photon and the electron. 

It was at this point that de Broglie suggested a new concept 
which is best expressed in his own words: 

"A consideration of these problems led me, in 1923, to the 
conviction that in the Theory of Matter as in the Theory 
of Radiation, it was essential to consider corpuscles and 
waves simultaneously, if it were desired to reach a single 
theory, permitting of the simultaneous interpretation of the 
properties of Light and those of Matter. It then becomes 
clear at once that, in order to predict the motion of a cor- 
puscle, it is necessary to construct a new mechanics, a theory 
closely related to that dealing with wave phenomena, and one 
in which the motion of a corpuscle is inferred from the 



motion in space of a wave. In this way there will be, for 
example, light corpuscles, photons, but their motion will be 
connected with that of Fresnel's waves, and will provide an 
explanation of the phenomena of interference and diffrac- 
tion. Meanwhile it will no longer be possible to consider the 
material corpuscles or electrons as discrete isolated entities; it 
will, on the contrary, have to be assumed in each case that 
they are accompanied by a wave which is bound up with 
their own motion. I have even been able to state in advance 
the wave length of the associated wave belonging to an elec- 
tron having a given velocity." 

We recall that in talking about sound waves, we found a 
simple relation (v = vX) between the velocity, v, the wave 
length, A,, and the frequency, v. Now, if the energy of the elec- 
tron is also directly proportional to the frequency, we should 
expect a simple relation between the velocity of the electron and 
the wave length to be ascribed to it. This relation is expressed 

, , Planck's constant, h 

by: Ihe wave length, k=— — 7- r—~. — : 7— 

J ° Mass of electron (m) Xits velocity, (v) 




Experimental Confirmation of de Broglie's Theory 

The theory outlined in the last section was mere theory . 
The important question then was: "Would electrons, or a beam 
of electrons, exhibit experimentally that there really is a wave 
motion associated with an electron, or, in other words, would a 
beam of electrons show the phenomena of interference and dif- 
fraction, the sine qua non of wave motion?" 

After de Broglie's announcement, this question was investi- 
gated and soon answered in the affirmative by two groups of 
workers, quite independently: by Davisson and Germer (1927) 
in the research laboratories of the Bell Telephone Company, 
New York, and by G. P. Thomson in work done at Aberdeen 
(1928). We do not need to go into the details of these experi- 
ments, but it will suffice to say that they answered the question 
decisively. A moving electron has a wave associated with it. 


The Wave Length Possessed by an Electron 

As noted above, the theoretical value of the wave length that 
de Broglie worked out for an electron is 


*= > 


The value of m, the mass of the electron, a constant for all elec- 
trons, was determined first by Sir J. J. Thomson and many times 
since, so both h and m are well determined constants. 

As we shall see later, the velocity of an electron can be easily 
determined if we know the accelerating voltage (V). Substi- 
tuting the various experimentally determined values in the above 
expression, it has been found that 

\= — — X 10" 8 cm. 


where V is expressed in volts. If the voltage is of the order of 
hundreds of volts, the wave length is of the order of 10 8 cm or less. 

In view of this fact, scientists interested in the microscope at 
once asked: "Can the electron waves be used in any kind of 
microscope?" Because, if they could, there would be promise of 
increasing the resolving power a thousand times or more. 

We know that the light microscope permits us to see particles 
l/125,000th of an inch in diameter. Smaller ones we cannot see. 
(Chap. 6.) We are limited by the wave length of the light. 
Using electron light, which has very much smaller wave lengths, 
we should be able to photograph very much smaller particles and 
also to reveal much greater detail in an object. 

This the electron microscope has accomplished. 

Electron Waves and Electron Rays 

In the development of his theory, de Broglie showed that the 
path of an electron, its so-called trajectory, as it moves through 
space bears much the same relation to its wave property as the 
ray of light, which we have used in our early figures, does to the 
wave property of light. In de Broglie's words, "the possible 
trajectories of the corpuscle are identical with the possible rays 



of its wave." Therefore, if we could determine these electron 
paths and treat them as "light rays," we might find out something 
about the possibility of focussing such rays. 

There are two ways in which the paths of electrons may be 
affected: (a) by electrical fields of force and (b) by magnetic 
fields of force. Of course, these electron beams cannot pass 
through material lenses, and so if we wish to produce the refrac- 
tion of electron-waves, we shall have to construct electrostatic 
or magnetic fields of force, which may serve as electrostatic lenses 
or magnetic lenses. All these operations must be carried out in 
a good vacuum. 

Courtesy American Cyanamid Co. 

Zinc oxide smoke ( X 30,000) . 





1 'I 


fxXXXvXv. v xx<;r ; x. \ x; \ ■ •• .- .xxx/-xxx* 

mimr--::-- ;xi^:., 


; xx :♦. ! ^pf. x v * *$Sr? '• - ' ' : ., x ;:; :, .-x, .■ ; 

Miff ''SL^- 1$*S* ' * ' ' ^ 

fe^;/;:^' * .,*•;. -.- x ' ' x 5 „x> :/ ;: -#lx? 

Courtesy RCA Laboratories 

Magnesium oxide smoke ( X 50,000) . 

Courtesy American Cyanamid Co. 

Tungsten oxide smoke (x 30,000). 



Courtesy American Cyanamid Co. 

Aluminum oxide smoke (X 30,000). 

Chapter 12 


Work, Force and Power 

We come now to a stage in the development of the subject 
where we shall have to be rather more precise in our language 
and to introduce some technical terms, which may need some 
elucidation. Some of these technical terms are simple words in 
everyday use, which, when used in a strictly scientific sense, need 
special definition. Such words are force, work and power. In 
our colloquial use of these terms, we may say a man is "a man of 
great force" or "a man of great power" and mean the same thing, 
and probably about the same as to describe the individual as "one 
who does an enormous amount of work." But when we come to 
use the term work, for example, in the scientific sense, we think 
of such a thing as carrying a load up a hill; this is exerting a force 
against the force of gravity. 

For example, suppose a crate weighing a ton has to be delivered 
at a factory platform (Fig. 53). It may be hauled up by a block 
and tackle or dragged up an inclined plane on rollers. It is almost 
self-evident that the work done or effort expended will be pro- 
portional to the pull one has to exert and also to the distance 
through which we have to pull the crate. It is a matter of com- 
mon knowledge that we do not have to exert as large a pull if we 
make use of the inclined plane as we do if we pull the weight 
straight up. When we make exact measurements of the pull and 
the distance in the two cases, it comes out that the product of the 
pull and the distance over which the object pulled moves is always 
the same for the same height of platform. If we use different 
lengths of inclines, i.e., inclines making different angles with the 
ground, we still get the constant value when we multiply pull by 




This measure of the work done is just the weight (2000 lbs) 
multiplied by the height (20 feet). What we called the pull is, 
in technical terms, called the force exerted; we then have the 
relation: work done equals force exerted times distance through 
which the force acts, or in mathematical shorthand simply: 

W = f x d 

In Fig. 53, f 1 Xd 1 = f 2 X d 2 = 2000x20 = 40,000 units. 

Fig. 53. The work done in pulling the load from the ground to the platform is 
the same no matter how the work is performed. 

We have mentioned also power; technically this is just the 
rate at which work is done. The relation is: 

Power (P) 

Work done 
Time taken to do it 




If in Fig. 53 we pull the rope of the block and tackle by hand, 
we take a long time to pull the weight up, but if we use a donkey 
engine of a few horsepower, we can haul up the heavy load in 
seconds, simply because the power of the engine is greater than 
the power of one man. 


Man's first ideas about electrical charges were that when 
bodies such as glass or ebonite were put in a state defined by say- 
ing they were charged, such bodies exerted a force on each other. 
As soon as we begin to talk about one charged body exerting a 
force on another, which is indicated by the bodies being pulled 
toward or repelled from each other (Fig. 42, p. 82), we think 
of gravitational forces — the attraction of the earth for bodies 
free to fall. However, with the electrical forces, we have not 
only attractjon but repulsion. In any case we come to the ideas 
of forces moving masses, or masses being moved against forces, 
i.e., work jDeing done; and in the end we have the relations 
between force, work and power 


W = f X d, and P = 

1 t 

as in the ordinary experiences in the lifting of weights. 

In these days we are all familiar with the development of 
"electrical power," as we say, by harnessing huge waterfalls. 
When the engineer wishes to calculate the rate of production of 
energy he uses just the same relation. 

Work done in any time equals the weight of the water fallen, 
f, times distance fallen, d, i.e., W = fxd, since the total weight 
of the water is just the force of gravity on the water. If the unit 
of time taken is one second, then the power, P, available at the 
falls will be: 

p= W fxd 

t t 

where t is the interval of time in seconds. 

Difference of Level and Difference of Potential 

When an engineer discovers a new falls or creates a difference 
of level by a dam, he knows that the power available depends on 
two things, the quantity of water falling down per second and 
the height of the falls (or dam) . Of course, fundamentally, he 
measures the height directly by the use of a yardstick or steel tape 
measure. But there is another perfectly good way to determine 
the height, namely, by measuring the work done in hauling up 


from the level of the bottom of the falls to the level of the top of 
the falls a given weight of water, say a pailful. 

Here we have just the concept of difference of level between 
two points, say A and B, on the earth; there are two things that a 
given difference of level between the points A and B will tell us: 
(1) in what direction water will flow, whether from A to B or 
from B to A, i.e., we must know which point is at the higher level; 
and (2) how much work is done in lifting a definite weight from 
one point to the other. All of these terms have been adopted in 
the language applied to electricity. 

If we have two points, A and B, either in space or on a con- 
ductor, say a wire, and there are electrically charged bodies 
around, which, of course, give rise to electrical forces, then there 
will be an electrical force acting between A and B and electricity 
will be moved from A to B (or B to A). This case is more 
complicated because we have two kinds of electrical charge to 
talk about, positive and negative, and any electrical forces will 
tend to make positive electricity move in one direction and nega- 
tive electricity move in the opposite direction. So we have to 
choose one or the other as a reference term; a certain charge of 
positive electricity has been chosen as the standard test quantity; 
that is, it takes the place of the pail of water used to measure 
differences of level. 

What we call difference of level in dealing with water falls 
is called in electrical language difference of potential, this is 
measured in volts. If we know the difference of potential between 
two points, A and B, we know which way positive (as well as 
negative) electricity will move, whether from A to B or B to A, 
and we know how much work will be done in taking a unit posi- 
tive charge from one point to the other. 

How are we to measure difference of potential? We have 
no yardstick or measuring tape for it. The only way it is meas- 
ured is by finding the work done in taking a certain unit quantity 
of electricity from one point to the other. In fact, the difference 
of potential between two points is defined as the work done in 
taking a unit of positive electricity from one point to the other. 


In the case of lifting weights, as in Fig. 53, the work done 
in lifting a given weight depends only on the difference of level 
between the ground and the platform and not on the path by 
which the weight is pulled (i.e., straight up or up along an in- 
cline) ; so the work done in moving a quantity of electricity from 
one point to another depends only on the difference of potential 
between the two points and not on the path along which the 
electricity is moved. 

Lines of Force and Equipotential Lines 

Fig. 54 illustrates a specific experiment in which the electrical 
terms introduced above are used. Two parallel metal plates, A 
and B, are connected to the ends of a battery so that A bears a 
positive charge and B a negative charge. If we put our unit 
positive test charge on a small pith ball, P, it will be pushed away 
from A and attracted toward B. If we have the experiment care- 
fully arranged, P will move in the direction from A to B along 
lines perpendicular to A and B, no matter where it is placed 
originally in the space between A and B. These lines then indi- 
cate at any point the direction in which the electrical forces are 
acting. They are called lines of force. 

Now mark off on these lines points corresponding to equal 
divisions of the total work done in taking P from A to B along 
any one of the lines of force reaching from A to B. Four such 
divisions are indicated by the dots. If we join the corresponding 

d cm 



!+! f : ! 

+ Ma 

;p ; : 



• — *■ — 

E, - 



Ba ttery 
Fig. 54. Lines of force and lines of equipotential. 


dots on the various horizontal lines by broken lines, we get a 
series of lines which mark off regions of equal work done in 
going out from A; these vertical lines are known as equipotential 
lines. This is a simple case, and it is apparent that the lines of 
force are perpendicular to the equipotential lines wherever they 
cross one another. 

From the symmetry of the lay-out in Fig. 54, we would 
easily agree to the statement that the force, f, acting on the small 
charged pith ball, P, will be the same at every point between A 
and B. From the expression giving the relation between work, 
force and distance, we can then say that: 

W (Work done in taking P from A to B) = f X d. 

But the definition of the difference of potential between A and B 
is just this work if the charge on P is a unit charge. So if we put 
the potential of A equal to V A and that of B equal to V B and indi- 
cate the force on unit charge by F, we obtain: 

W = V A - K B , and so V A -V B ='Fxd 

From this equation, we get a very useful relation F = — -• 


Now if everything is symmetrical as in Fig. 54, — — is the 


fall of potential per cm., which is consequently called the poten- 
tial gradient from A to B. We then arrive at the statement that 
the electric force per unit charge at any point is equal to the 
potential gradient, sometimes called the gradient of the potential. 
This quantity — the force per unit charge at any point — is also 
called the intensity of the electrical field and is usually denoted by 
the capital letter, F; this is sometimes called the field strength. 
It is the difference of potential between A and B divided by the 
distance between the two plates. 

Again, since by definition the difference of potential between 
two points is the work done in moving a unit positive charge from 
A to B, we can readily deduce that twice as much work will be 
done in moving two units of positive charge from A to B and 
9 times the original work when a charge of 9 units is moved from 


A to B. We can express this in a general way by the relation: 
W = qX(V A —V B ), if the charge is q units. 


W = fxd where f is the electrical force acting on the 
charge, q, we may write 


fxd=qx(V A 

V B ). 


= F, it follows that f = q X F, or force 

equals charge times field intensity. 

If a positive charge q moves from the positive plate A to the 
negative plate B in Fig. 54, it moves in the direction of the electric 
field F, and we say that the field does the amount of work, W, on 
the charge q. This work appears as the kinetic energy, \mv 2 ^ 
which the charged particle of mass, m, possesses when it arrives 
with the velocity v, at the plate B. 

On the other hand, if we were to move a positive charge q 
from plate B to plate A, the charged particle would have the 
work W spent upon it. Either we must push the particle along 
against the force of the field or the particle must possess an initial 
energy equal to W in order to be able to sacrifice that much on 
its way from B to A. This can be realized by the following 
experiment (Fig. 55). 

We arrange a third plate, C, behind the plate B at the same 
distance, d, which separates A and B and connect another battery, 
F 2 , of the same voltage as E u so that plate C is positive with 
respect to plate B by the same amount that plate A is positive with 


Fig. 55. Gain and loss of energy of a charge in an electric field. 



respect to B. The fields F u F 2 , in the space to the left of B and 
the right of B will then be equal and opposite. If we pro- 
vide a small hole in B and let the positive charge originate at C, 
it will acquire the energy W while traveling from C to B and then 
lose it while travelling from B to A. It will arrive with zero 
velocity at A. 

The same reasoning applies naturally to negative charges. If 
e stands for the negative charge of an electron, the work done on 
the electron by moving it from plate B to plate A in Fig. 54 is 

W = ex(V A -V B ) 

and if it were to be moved from A to B, it must have this initial 
energy in order to be able to overcome the opposing field force. 
That could be realized by an experiment similar to that illustrated 
in Fig. 55. When a field does work on the electron and thus 
accelerates it, the energy gained by the electron is expressed fre- 
quently in terms of electron-volts. Thus an electron, which, 
starting from rest, has been accelerated by moving between two 
plates having a difference of potential of 50 volts, is said to gain 
a final kinetic energy of 50 electron-volts of energy. It is also 
sometimes said to have a final velocity of 50 electron-volts. There 
are then unfortunately two kinds of electron-volts in use, one 
referring to the energy of an electron, the other to its velocity. 

Fig. 56 represents a more general case of lines of force and 
corresponding lines of equipotential. 

We have now to deal with the motions of electrons in vacuum 
tubes under the influence of electrical forces brought into action 
by maintaining two metal plates in the vacuum tube at a difference 
of potential. 


Lines of [qui 'potential 

V * A 

Lines of Force. 

> « « 


Fig. 56. 

The field of force 
between two spherical 
conductors at a differ- 
ence of potential. 


The Motion of Electrons in a Uniform Field 

The electron stream observed in the earliest experimental 
tubes, such as the Crookes tube and the Braun tube, shown dia- 
grammatically in Fig. 57, was produced by the impact of positive 
ions on the negative electrode. 

As we have noted above (p. 97), these impacts of the positive 
ion masses will knock electrons out of the plate, C, really in great 
profusion, and these electrons will be forced away from C toward 
A along one of the lines of force. 

The resulting motion of the electron will be quite similar 
to the motion of a stone when we drop it to earth. We know 
that as soon as the stone is released it begins to fall with a velocity 
which becomes greater and greater the longer the stone is falling. 
This is what is called technically "accelerated motion." 

Using this analogy we might say that the electron, P, (Fig. 
57) "falls" from C to A with a uniformly accelerated motion. 
As the final velocity with which the stone reaches the earth 
depends on the height from which it has fallen, so the final veloc- 
ity of the electron depends on the difference of potential through 
which it has fallen. The greater the potential difference V A — V G , 
the greater is the final velocity. 

From what has preceded we see that we always use the term, 
difference of potential between points, and do not speak of the 
actual value of anything called potential at any point without 
referring to an assumed reference potential. This is quite in 
keeping with what the engineer talks about when he is calculating 
the power that he can get out of a waterfall. Niagara Falls has 
a height of 190 feet and the available power is fixed by this differ- 
ence of level but the engineer in making his calculation does not 

Fig. 57. 

The motion of an 

electron in a uniform 

field in a vacuum tube. 

90 volts 


need to know how far the bottom or top of Niagara Falls is above 
sea level. In fact, sea level is just an adopted level from which 
heights of individual stations may be measured. Similarly, in 
electrical measurements, the earth, which is a huge conductor, is 
said to be at zero potential, and this fixes our reference point. 

Thus the potential of a conductor joined electrically to the 
earth, or as we say, to ground, is said to be zero. With this defi- 
nition of a standard potential, called zero potential, a positively 
charged body is said to have a potential above the earth, because 
positive electricity will flow from such a body to the earth; a 
negatively charged body is one which is said to be below the 
potential of the earth, or at a negative potential, because positive 
electricity will flow from the earth to such a body. 

Referring to Fig. 57, we may say that the plate A is 90 volts 
above ground (or +90V) when it is connected with the positive 
terminal of the battery, the negative terminal of which is con- 
nected to a water main, giving good electrical connection with 
the ground (the earth). If the positive terminal of the battery 
is grounded, then the plate, C, is said to be 90 volts below ground 
(or —90V). The essential point to remember is that we use the 
term potential only in the sense of difference of potential. 

Once more referring to Fig. 57, we may speak of the potential 
at a point in free space, such as X. This can be found by measur- 
ing the amount of work that must be done in order to bring a 
unit positive charge from plate C to X. If the point X were to 
lie on the plate C, we would not have to perform any work, since 
the potential V G is the same at any point of C. If the point X, on 
the other hand, were to lie on the plate A, we would have to move 
a unit positive charge from C through space to A and overcome 
the repelling electrostatic force all the way until we reach the 
potential of +90 volts on A. By actually measuring the amount 
of work performed, it can be shown that it rises linearly all the 
way from C to A. If the point X lies half way between A and C, 
the potential at X is then 45 volts above C. 

The Paths of Electrons in Non-Uniform Electric Fields 

We shall now consider a case where the equipotential areas 
in space between cathode and anode are not plane but curved, at 


least in part. This may be realized by giving the cathode a 
spherical shape. We know that the surface ot a conductor is 
everywhere at the same potential. If the cathode is grounded, 
the zero-potential area then must of necessity follow the contour 
of the cathode (Fig. 58) . The other electrode is again connected 
to a 90-volt battery and the potential values at the various equi- 
potential areas are indicated. Since the anode is a plane, the 
equipotential planes gradually straighten out the nearer we ap- 
proach the anode A. 

We now assume that electrons are emitted with zero initial 
velocity all over the cathode surface, as indicated by the small 
dots. From each such electron, a line is drawn toward the anode 
in such a way that it intersects all equipotential planes at right 
angles. We know that this is the path the electrons must follow 
according to rule. The result is that all electrons meet at a small 
circle at the anode just as if they had been focussed. 

Fig. 58. 

Lines of force and 
lines of equipotential in 
a non-uniform electric 

This sort of thing had been observed by Crookes, Hittorf, 
Goldstein and others during their early experiments, and they 
contrived many different shapes for cathodes and anodes in order 
to learn more about the paths. We must remember, however, 
that these scientists did not yet know that they were dealing with 
electrons and they also had no clear conception of potential dis- 
tribution and its importance for the determination of the paths or 
trajectories of the emitted particles. 

We notice in Fig. 58 that the electrons do not travel in 
straight lines when the equipotential areas are curved, i.e., when 
the field is not uniform. In analogy with the reflection of light 
at a concave mirror one is tempted to speak of an electron focus. 


Courtesy University of Toronto 

Hairs on edge of midge's wing (Chironomidae) (X 20,000). Insert: Hairs on surface. 

Chapter 13 


Magnets and Magnetic Fields 

It has been mentioned on a few occasions in the preceding 
text that the paths of electrons may be influenced by the 
proximity of magnets. Thus, the fluorescent spot on the wille- 
mite screen of a Braun tube can be shifted at will when a small 
bar magnet is brought near the tube and moved into different 
positions. Whenever a mass, no matter how small, is deflected 
from its original path, a force must be acting upon it; the force is 
the cause of such a deflection. We know of mechanical forces, 
gravitational forces, and electrical forces. Now we must add 
magnetic forces to this list and learn something about their inter- 
action with electron beams, since this interplay is important in the 
operation of the electron microscope. 

Everyone is familiar with the ordinary magnetic compass — a 
small piece of steel pivoted so as to move in a horizontal plane. 
Its usefulness depends on the fact that one particular end always 
points toward the north, when the compass is kept at rest and 


Loclestone Compas5 


Fig. 59. Lodestone and compass needles. 



level. The end which points to the north is called the north pole, 
that pointing to the south, the south pole (Fig. 59). 

It is really not known exactly when compasses were first used, 
but the earliest forms use a magnetic substance which occurs 
naturally on the earth's surface. This substance called lodestone 
(leading-stone) is made up of oxides of iron. If an irregular piece 
of this mineral is suspended by a thread so as to be balanced hori- 
zontally, the longest dimension will always set so as to point in 
a north-south direction — one definite end always pointing to the 

Fig. 60. 

The laws of magnets; 
unlike poles attract each 
other, like poles repel 
each other. 

The question may be asked, "How did the modern small steel 
magnet develop from the lodestone?" This leads us to speak of 
a very important property of lodestone, i.e., that it is able to 
attract to itself pieces of iron or substances containing iron. Not 
only will the lodestone attract iron, but if it comes in contact with 
a piece of iron, it will impart its own power to the iron and so con- 
vert it into a magnet. Every boy who owns a jack-knife knows 
that he can magnetize it by rubbing it on a magnet. All our 
magnets are made of steel, because steel will retain this curious 
magnetic property longer than soft iron. 


When people began to experiment with these magnetic 
needles, they discovered a very curious property which has to do 
with the interaction of one magnet on another. If we have two 
magnetic needles and pick one up and bring one end up to the 
end of a second magnet, we find that two poles which are differ- 
ent (N and S) always attract each other, but that two poles of 
the same kind (N and N, or S and S) always will repel each other 
(Fig. 60.) This last fact recalls to us something similar regard- 
ing electrical charges — like charges repel, unlike charges attract 
each other— but this does not prove any real connection between 
magnets and electrical charges. 











M 5 


M" 5 " 

Cl— "1 


.... Sj 

«& 1 

Fig. 61. Breaking up a magnet into many parts always makes a complete magnet 

out of each part. 

One very remarkable difference is apparent at once. We may 
impart to a body a positive or a negative charge and detect its 
presence by simple tests. But we cannot detect both a negative 
and a positive charge on an isolated conducting surface, since the 
two charges neutralize each other. For magnets, the opposite 
condition exists. Each magnet always has a north pole and a 
south pole and we can never get any separate portion of a mag- 
netic substance with only one kind of pole, north or south. If we 
have a piece of steel which is magnetized and break it into any 
number of parts, all the parts are still perfect magnets with both 
north and south poles (Fig. 61). 



Since one magnet will exert a directive force on another, and 
since any magnet always sets itself in a north-south direction 
when free to do so, we decide that the earth itself must have a 
directive action on a magnet. From this we conclude that the 
earth itself is a huge magnet also having two poles, i.e., points 
toward which the north and the south poles of magnets are 
directed. The magnetic pole in the northern hemisphere is in 

Fig. 62. 

The magnetic lines of 
force about a bar mag- 

northern Canada near Hudson Bay; this point is known as the 
north magnetic pole. Neither of the earth's magnetic poles are 
in the same place as the geographical north and south poles. 

Just as in the case of the attractions and repulsions between 
electrical charges, so we have here the remarkable fact that one 
body affects another apparently "at a distance." In this case also 
Faraday ascribed this interaction to something going on in the 
space round about the magnets. We speak, then, of the region 
round about a magnet as a magnetic field of force, and also of 
magnetic lines of force which show us graphically the direction 
of the magnetic force at any point round about any magnet. 
These magnetic lines of force can be easily found in a plane by 
moving a small compass around on a piece of cardboard placed 
over the magnet and marking at every point the direction in 
which the compass needle sets itself. 

Fig. 62 shows the distribution of lines of force about a bar 
magnet and Fig. 63 that between the poles of a horseshoe magnet; 
the remarkable and, as we shall find later, useful property of the 
field of the horseshoe magnet is that the most intense force is in 


the region between the poles and that in the central part of this 
region the field is uniform, i.e., the force on a certain specially- 
defined unit pole is the same at every point. 

It has been agreed to define the direction of the magnetic line 
of force as the direction in which a north pole tends to point. 
This is the direction indicated by the arrows in Fig. 62. 

Fig. 63. 

The magnetic lines of force for 
a horseshoe magnet. 

The Magnetic Field about a Conductor bearing a Current of 
Electricity: The Electromagnet 

What has a magnet to do with an electron? The answer 
to this question goes back to an accidental discovery in 1820 by 
a Danish scientist, Oersted, that an electric current flowing 
through a wire exerted a magnetic effect in the region round 
about the wire. In other words, the current through the wire 
set up a magnetic field. This is illustrated in Fig. 64, where a 
horizontal wire, which is part of an electrical circuit, is shown 
with magnetic compass needles pivoted above and below the wire. 
As represented, the horizontal wire is placed so as to lie in a north- 
south direction; when no current is flowing (Fig. 64 A), the 
needles both point due north. When the switch is closed and 
the current flows in the wire in the direction determined by the 
position of the plus and minus signs on the battery, the north poles 
of the needles move in the directions indicated by the arrows. 
If we follow the movement of the poles of the magnet in its 
various positions, we are led to the conclusion that the magnetic 



iYo current ftousincj 

Current flom-inc/ 

Fig. 64. The direction of the magnetic lines of force about a conductor through 
which a current is flowing. 

lines of force about the wire are circles in planes at right angles 
to the wire and with their centers in the wire. The relation 
between the direction of the current and the direction of the 
magnetic lines of force is given by a very simple rule (Fig. 65) . 
If we are using a screwdriver to drive in a screw, and let the direc- 
tion of progress of the screw represent the direction of the 
current, the direction of rotation of the thumb represents the 
direction of the magnetic lines of force. 

^ fcmA 



Direction of 
magnetic fine ot force 

Direction ot current 

Fig. 65. A simple rule for remembering the relation between the direction in which 
a current is flowing in a conductor and the direction of the magnetic field 
set up. 


Now if we bend the wire of Fig. 64 into a circle, as in Fig. 66a, 
the lines of force, if we could see them, would appear to come out 
of the inner plane of the circle, bend around the wire and re-enter 
the other face of the circle of wire. This is shown more clearly 
by viewing the circle of wire edgewise from the left, as in 
Fig. 66b. In fact this figure suggests that the circle of wire acts 
as though it were a very thin flat magnet, such as we might get if 
we sliced out a very thin section of the bar magnet shown in 
Fig. 62. 

" (b) 

Fig. 66. 

The direction of the 
magnetic lines of force 
about a conductor bent 
in a circle and the de- 
velopment of the elec- 

So/ en o i d 


If, instead of a circle of one turn, we have a long wire coiled 
up in what is called a solenoid (Fig. 66c), we have an exact 
parallel of a long bar magnet; such a solenoid is known as an 
electromagnet. One end of the solenoid is a north pole and the 
other end a south pole, the polarity of either end depending on 
the direction of the current through the coil. 


In this case we have produced the equivalent of a bar magnet 
without the use of any lodestone, iron or any other magnetic 
substance. One very important property of such an electro- 
magnet is that the strength of the magnetic field within the coil 
of wire is uniform (see Fig. 67). However, we can get a much 


Fig. 67. The magnetic field at the center of a long solenoid is uniform. 

stronger magnetic field about the electromagnet if we insert a 
core of iron or steel in the solenoid. 

The great usefulness of the electromagnet is due partly to the 
fact that by increasing the value of the current through the wire, 
we can increase the strength of the magnetic field over a great 
range of values. Again, by employing an alternating current, it 
is possible to cause a corresponding alternation in the direction of 
the magnetic force. 

The Action of a Magnet on a Conductor Bearing a Current 

Oersted's discovery showed that a current in a conductor can 
move a magnetic needle: that is, a fixed conductor bearing a cur- 
rent affects a movable magnet. The converse of this is that a 
fixed magnet ought to produce motion of a movable conductor 
which is bearing a current. This can be shown to take place by 
a simple experiment (Fig. 68) . A battery sends a current through 


a wire, AB, which is suspended from a metal hook, A, and dips 
into a pool of mercury, M; a bar magnet is supported through a 
cork at the bottom of the tube. When the current is sent through 
AB, the lower end rotates about the pole of the magnet as long 
as the current flows. Here the motion of the conductor is per- 
pendicular to both the direction of the current and the direction 
of the magnetic lines of force. 

Fig. 68. 

A stationary magnet can 
produce motion in a conductor 
bearing a current; the converse 
of the action shown in Fig. 64. 

Now if electrons are negatively charged particles, a stream 
of electrons moving along a straight line should be equivalent to 
a flow of negative electricity in that direction. Consequently, if 
the stream of electrons is sent through a magnetic field, the mag- 
netic force should move the electrons in the stream in a direction 
at right angles to both the direction of the electron stream and 
the direction of the magnetic field (Fig. 69). 

It was just this observation which led Sir J. J. Thomson to 
conclude that the electron stream acted as though it were a 



f/ectron stream 


f''*fe*!9**i? c 




Fig. 69. The deflection produced in the path of an electron passing through a 

magnetic field. 

stream of negative electricity and to announce that electrons were 
small particles of matter charged negatively. 

Experiments show that the electron is affected by a magnetic 
field only when it is moving. If the electron is at rest, there is 
no action whatever. 

Comparison of the Action of Electric and Magnetic Fields on 

We find then that the flight of electrons is affected by the 
presence of a magnetic field as well as by the presence of an elec- 
trical field. It is important to contrast the effects of these fields. 

An Electrical Field 

1. Acts on an electron which is 
initially at rest and causes it to 
move along the lines of force 
in a direction opposite to the 
motion of a positive charge. 

2. Acts on moving electrons in 
such a way as to bend the 
beam into regions of higher 
potential (Fig. 58). This re- 
fraction of the electron beam 
at the equipotential boundary 
layers is accompanied by a 
variation in the speed of the 

A Magnetic. Field 

1. Has no effect whatever on an 
electron which is at rest. 

2. (a) Has no effect on a moving 
electron as long as the direc- 
tion of motion is along the 
lines of force of the magnetic 

(b) Moves the electron in a 
circular path, if the direction 
of motion of the electron is at 
right angles to the lines of 
force or the magnetic field. 
This circular path is still per- 
pendicular to the magnetic 
lines and the speed of the elec- 
tron is not altered. 


Fig. 70a illustrates what will happen to a moving electron 
entering a uniform magnetic field in a direction at right angles 
to the lines of force. Immediately the electrons enter the mag- 
netic field, a force is exerted on them which tends to bend the 

Fig. 70. The path of an electron entering a magnetic field in a direction at right 
angles to the lines of force is changed to a circle. Fig. 70b gives an end-on 
view of the circular path; the dots represent the sections of the lines of force. 

electron beam into a circle of definite radius; this circular motion 
persists as long as the electron remains in the field. The value 
of this radius, R, varies directly as the velocity of the electrons, 
and inversely as the intensity of the magnetic field. The technical 


expression is R =— 9 where m is the mass of an electron, v its 

velocity, e its charge, and H the value of the magnetic field; these 
values must be expressed in terms of definite units in order to get 
the value of R in centimeters. 

If the magnetic field occupies a small space, the electron may 
escape from the region of the field before it completes the circular 
path; this is illustrated in Fig. 71. The electron beam enters the 
field at A and escapes at B; while moving in the field, the electrons 
follow the path AB, which is an arc of a circle with the radius 
given by the above relation. That is, AB is a part of the dotted 
circle, which would really be the path of the electron if the 
uniform magnetic field extended farther in space. 

At the point B, where the electrons leave the magnetic field, 
the action of the magnetic force ceases and the electrons continue 
their flight in a straight line along the tangent of the circle at B. 
The circle which is indicated by the dotted line in Fig. 71 



/ \ 
/ ^ 




1 j 




i _ / 
- t -_ R _7 — _- - 

Fig. 71. The path of an electron which escapes from the magnetic fie 


appears as an ellipse since it is drawn in perspective. Figs. 70b and 
72 give the electron trajectories which an observer would see 
while facing the magnetic field. The dots indicate the field 
arrows. This principle of magnetic deflection of electron beams 
in the manner just described is used very generally in cathode-ray 
tubes and television tubes for scanning. 

Fig. 73 illustrates this application of magnetic deflection in 
its simplest form. Two solenoids, Ci and C 2 , are placed, as 
shown, at right angles to the axis of the cathode ray tube. This 
is a modern Braun tube in which an oxide-coated cathode is used 
as the source of electrons; many other refinements have been 
incorporated. A sharp spot of light is formed on the luminescent 

Fig. 72. 

End-on view of Fig. 
72. The dots represent 
sections of the magnetic 
lines of force. 


screen, S, at the end of the tube. When the coils are activated 
by an alternating current, an oscillating magnetic field will be 
established along the axis A-B of the coils. Since the field, H, is 
proportional to the current flowing through the coils of the 
solenoids, it will reverse its direction continuously according to 
the alternations of the current, i.e., H will be directed from A to B 
at one moment and from B to A l/60th of a second later, if 
60-cycle current is used. The effect of this alternating field, H, 
will be that the fluorescent spot on the screen, S, describes a 
vertical line flitting back and forth between P x and P 2 . 

Fig. 73. The magnetic deflection of an electron stream due to an alternating field, 

Path of an Electron Moving in any Given Direction on Entering 
a Magnetic Field: Vectors 

Let us now investigate what takes place when the magnetic 
field and the electron beam do not happen to be at right angles, 
as in the previous examples, but are inclined to each other at a 
smaller angle. 

For this purpose we need to say a few words about directed 
quantities in physics, which are called vector quantities. As 
examples we mention force, velocity, field strength; each of these 
quantities is determined by a numerical value and a direction. 


The velocity, or speed, of a moving object for instance, may be 
50 miles per hour where 50 is the numerical quantity. In order 
to describe the movement fully, we must also indicate the direc- 
tion in relation to some reference point. Thus we say, a car is 
travelling north on a highway with a speed of 50 miles per hour. 
Everybody then knows exactly what the car is doing. Other 
quantities in physics are fully described by a mere number. They 
are called scalars; examples of scalar quantities are mass and tem- 
perature; the concept of direction is not associated with these 
terms. It has become common practice to call the scalar part of 
the velocity by the word speed. 

Vector quantities are generally represented by arrows where 
the length of the arrow indicates the numerical value and the 
arrow head, the direction; the arrow is called a vector — merely 
a straight line having a definite length and a definite direction. 

A F 2 B 

^ < a o — >7 

* Fig. 74. 

* The sum of two vectors. 

• ° 


The interesting feature of these vectors is that they can be 
split up graphically into various components or, what amounts 
to the same thing, several vectors can be added by simple rules 
in a diagram. Let us illustrate this rule for a force diagram. 

In Fig. 74, a force F 1 of 3 units, acts upon a mass situated at 
O in the direction O-A for a certain time so that the mass is moved 
from O to A. A second force, F 2 , of 4 units is then applied for 
the same time at A in the direction A-B so that the mass is moved 
to B. The same result could have been obtained by having a 
force, F 3 , equal to 5 units act on the mass at O in the direction 
O-B for the same time. As the vector F 3 accomplishes the same 
effect as F x and F 2 both acting together, F 3 is said to be the sum 


of Fi and F 2 . We could also have applied F 2 first to get to C 
(Fig. 75a) and then Fi to get from C to B and would have ob- 
tained the same result. By drawing the two arrows F u F 2 from 
the common origin O (Fig. 75b), we find that a rectangle results 
of which F 3 is a diagonal. If the two original forces act at a differ- 


F 2 


Fig. 75. The sum of two vectors. 

ent angle with respect to each other, as in Fig. 75c, we obtain a 
parallelogram. In every case, the diagonal of the rectangle, or 
parallelogram, obtained in this graphical manner gives the magni- 
tude and direction of the force F 3 which brings about the same 
final displacement as that produced, in steps, by F 1 and F 2 . This 
is the well known parallelogram of forces probably familiar to the 
reader from his school days. Instead of forces, the arrows may 
represent velocities or other vector quantities. 

o o 

(a) (b) (c) 

Fig. 76. A vector split into two components. 

By reversing the vector addition, we find that we can split up 
a vector into two components that add up to the original vector 
as long as the original vector forms the diagonal in a rectangle 
or parallelogram. 

If the arrow marked v in Fig. 76a represents the magnitude 
and direction of a velocity with which a body moves from O, 
we will get to B by moving in two steps, v x -\-v 2 , as shown in 
Figs. 76b, and 76c. 


After this little excursion we now return to our main issue, 
i.e., the influence of a magnetic field on moving electrons. In 
Fig. 77, we picture an electron moving with a velocity v in a uni- 
form magnetic field H, so that the direction of v and that of H 
make an acute angle. We have no rule that would enable us to 
predict directly the trajectory which the electron will follow. 
But we do know the effect of H upon electrons moving parallel 
with, or at right angles to, H. Here we find the breaking up of a 
vector to be a welcome solution to our problem. We simply 
assume for the moment that the electron moves first with a 
velocity, v u at right angles to H and then with a velocity, v 2 , 
in the direction of H and deduce what path will result when both 
v 1 and v 2 are possessed by the electron at the same time. 

■*- H 

Fig. 77. The velocity of an electron in a magnetic field resolved into two new 
directions, one parallel to the field and the other at right angles to the field. 

It is quite plain from the rules which we have stated above that 
the electron moving with velocity and direction v ± in Fig. 77 
describes a complete circle which is oriented at right angles to 
the direction of H and thus returns to P from where it started. 
We are also certain that the electron moving with v 2 in the direc- 
tion of H is not affected by H and thus proceeds in a straight line. 
These two motions are represented in perspective in Fig. 78. 
Here a long solenoid is drawn which supplies the axial field, H. 
It has to be long in order to insure a uniform field in its central 
section where we shall study the electron movement through the 
cut-away shown. We assume that the electron has the velocity 
V represented by PB when it reaches the point P, on the solenoid 
axis O O', which of necessity will be the axis of a vacuum tube 


o - 


i... in- in- 
in- /i- in- 

Fig. 78. The focussing action of a magnetic field on an electron diverging from the 

direction of the field. 

inserted in the solenoid. What happened to the electron before 
it arrived at P does not concern us here as long as we assume that 
it had a linear motion in the direction PB. The moment it 
arrives at P the switch which actuates the solenoid is closed and 
produces the magnetic field instantly. Now we split v into v 1 
and v 2 and ask what path the electron will follow when it has 
these two velocities simultaneously. The electron tends to de- 
scribe the circle but at the saime time it progresses with the uni- 
form velocity, v 2 , toward the right. The path thus becomes a 
helix which is wound about a fictitious cylinder touching the axis 

Fig. 79. 

View of previous figure 



O-O'. Fig. 79 shows the view obtained by looking into the 
solenoid from the right. By the time the circle would have been 
completed the electron has arrived at P x . In the absence of a 
magnetic field, the path of the electron would be along the 
straight line to B. We thus make the important observation that 
an electron which tended to move away from the axis (in a 
divergent beam) has been forced back onto the axis some distance 
away, because of the action of the coaxial magneqc field. This 
coaxial field, then, acts upon a divergent beam of electrons in the 
same way as a lens acts on a divergent beam of light, i.e., it pro- 
duces a focus. Fig. 80 is the drawing of a space model which 
illustrates the trajectories of a number of electrons which originate 
at P with the same constant velocity but in different directions 

Fig. 80. 

n» The helical paths of 

electrons from source P 
to focus P\ 

with respect to the axis. They all converge toward F, the focus 
of the magnetic lens. 

The lens property of the long solenoid was recognized at an 
early date. E. Wiechert in 1899 introduced the concentration 
coil, as it was called, in connection with the Braun tube and it 
proved to be a most useful device in the early days of cathode 
ray tube technique. 

Edge of particle of pollen dust (X 21,000). 

'■•-.•"■ ' . • '■■■.'.. 


Courtesy Dr. Stuart Mud 

Vibrio comma (X 22,500) Cells dried from distilled water. 


* A 

Courtesy Dr. Stuart Mudd, University of Pennsylva 

Vibrio Comma, X 25,000, after brief exposure to a lead salt. 

Chapter 14 


The Magnetic Lens 

It may be well here to review the conditions that prevailed 
in the experiments just described which led us to ascribe a lens 
action to a solenoid. Electrons emerging from a small spot of the 
cold cathode of a Braun tube were accelerated in the close vicinity 
of the cathode surface by the cathode potential drop of the gas 
discharge column and then traveled with a fairly uniform velocity 
toward the fluorescent screen. During their travel these electrons 
would suffer collisions with gas molecules and be deflected from 
their original direction of travel thereby, so that on the whole a 
diffuse cathode stream would result. This is illustrated in Fig. 
81a, where the shaded part represents a cone filled with electron 
beams emerging from a point, P, on the cathode, K. The diameter 
of the fluorescent circle shown on the right of Fig. 81a would be 
correspondingly large, say several centimeters. It was necessary, 
for this reason, to insert an aperture disc, D, which served to 
block out most of the stray electron beams and permit passage 
only to a much narrower cone that could reach the screen, S. The 
diameter of the circle produced on the screen would then be 
much smaller, say a few millimeters; but the brightness of this 
luminous circle would be the same as that of the larger circle in 
Fig. 81a, since the same number of electrons impinges on the 
screen per unit area. 

When, instead of the insertion of the aperture disc, a solenoid, 
C, is slipped over the entire length of the tube and a current passed 
through its winding, the diameter of the luminous circle on the 
screen may turn out to be as large as in Fig. 81a, or nearly so. 
By gradually increasing the strength of the current flowing 




through the coil, by means of a rheostat inserted in the circuit, 
the large luminous circle first obtained will shrink and finally 
become a bright small spot, D 3 . The brightness will increase 
continuously since the total number of electrons reaching the 
screen in Fig. 81a will be concentrated into a smaller and smaller 
area on the screen. The spot of minimum diameter will have a 
dazzling brilliancy. When the current is further increased, the 
spot diameter will increase again to its old maximum, then de- 








P&\ D 


\ZJ \ z 



i-a -— f 


oooooooooo oooo o ooo 

Fig. 81. Magnetic lens in and out of focus. 

crease again, and so on. A great number of such cycles can be 
passed through. 

The variation of the diameter of the spot on the screen when 
the solenoid current is continuously increased brings to mind the 
similar variation in diameter of a light disc produced by a sun- 
glass on a piece of paper when the distance of the lens from the 
paper is varied. There is one position of the lens where we can 


produce a sharp focus on the screen S (Fig. 82a). The parallel 
rays from the sun are gathered by the glass lens, L, and combine 
in a brilliant spot on the screen, which may burn it. The distance 
from the plane of symmetry of the lens to the screen is then equal 
to the focal distance f and coincides with the image distance b. 
If the lens is moved to the left (Fig. 82b) the focal distance re- 
mains unchanged and the image distance b u is now larger than f . 


- \-::::::::X$&** 

- \ /••• '•• 


(b) — • 


• •• 

• ; • • .i r»'.v>* v**fcsw 


Fig. 82. Optical lens in and out of focus. 

The result is that the rays cross in front of the screen and diverge 
toward the screen to produce a diffuse image which is larger than 
that in Fig. 82a. If the lens is moved toward the right (Fig. 82c), 
a similar diffuse spot results, since the screen intersects the rays 
at a distance b 2 before they have come to a focus. Instead of 
moving the lens the screen, S, may be moved to the right or left, 
giving the same effects. The essential point to be emphasized in 


this optical experiment is that we can obtain only one focus. 
This is, of course, because the focal distance, f, of the lens is a 
fixed quantity which depends on the curvature of the lens surfaces, 
the thickness of the lens, and the type of glass used to make the 

When we come to describe the results obtained in the elec- 
trical experiment in which the coil current was continuously in- 
creased we arrive at Figs. 83a-e. The coil itself is not drawn in 
these figures but it is understood that a coaxial magnetic field is 
present in all five cases. Since we are dealing with a sealed-off 
vacuum tube the object-image distance, L, from cathode to screen 
remains constant. As a result of the increasing current intensity 
in the solenoid the focal length, f, of this magnetic lens evidently 
becomes shorter and shorter until in Fig. 83e its value has 
been halved. Since the electrons, after crossing the axis at F 4 , 
continue to be subject to the coaxial magnetic field, they are 
brought to a second focus, F/, on the screen. The long solenoid 
then acts in this particular case as a double lens. The variation of 
the spot diameter on the screen which is shown in Fig. 83 as D x 
. . . D 5 is now readily understood. 

It is apparent from this experiment that a magnetic lens ex- 
hibits the unusual property of having a focal length which can 
be varied at will simply by varying the amount of current which 
flows through its winding. This feature is of the greatest prac- 
tical importance. The magnetic field, H, represents the refrac- 
tive medium which acts upon the electron beams as the glass does 
on the light beams in the optical microscope, and it is the value 
of H which is increased when the current increases. 

The Short Magnetic Lens 

When we compare the space that is occupied by the refractive 
medium in the case of a glass lens with that of a long magnetic lens 
we realize that the glass lens is very much shorter. It would be 
very desirable to have the magnetic lens also contracted to a 
shorter length, because this would make its use much more flexible. 
During the early years of experimentation with the Braun tube, 
not much attention was given to the dimensions of the concentra- 



(a) k ^r_ — 


(b) K fc~ 

(c) K ll- r --e-->^"-^W 

(d) K ^-~><- 


Fig. 83. Varying the focal length of a magnetic lens. 

tion coil introduced by Wiechert. It was more or less fitted to 
the shape of the tube, sometimes extending over its full length, 
at other times fitted between cathode and aperture disk or be- 
tween aperture disk and screen. In 1905, Rankin found that a 
spot of greater definition and brilliance could be obtained by 
arranging a solenoid between cathode and aperture disk, i.e., 



between K and D in Fig. 84. On the basis of the lens action of 
this coil, we can now explain this phenomenon. In the absence 
of any coil, only a small fraction of the total electron stream 
emanating from the cathode penetrates the aperture of D (Fig. 
81b). A concentration coil, C, in the cathode-aperture disk 
region will focus all electrons in the aperture, thus making for a 
much brighter spot (Fig. 84). 


o o o o o oooooo 
Fig. 84. Concentration of electrons on an aperture. 

After the electrons leave the magnetic field at the aperture 
disk one would expect them to continue their flight along the 
dotted arrows in Fig. 84. In a gas-filled Braun tube they remain 
confined to the axis because of an effect known as gas concentra- 
tion. Positive ions created along the axis of the tube exert an 
electrostatic attraction upon electrons which tend to diffuse away 
from the axis. In a tube free of gas, a high-vacuum tube, the one 
coil applied in Fig. 84 would indeed be of no use, since a very 
large luminous circle would be obtained. This condition is 
readily rectified by applying two separate concentration coils, one 
to the left of the aperture disk, Ci, and one to the right, C 2 
(Fig. 85). 

oooooeooo . 

ooooooo oooooo 

Fig. 85. Use of second concentration coil. 

The first coil produces an image in the aperture plane and the 
second coil uses this image as a virtual object of which it produces 
in turn a sharp image on the screen, S. The size of the spot is now 


determined by the diameter of the aperture in the disk, D. This 
system was first realized by Rogowski and Flegler in 1927. 

The first demonstration of magnetic focussing was given in 
1896 by MacGregor-Morris, together with Professor Clinton, 
who, on the suggestion of Sir Ambrose Fleming, pursued experi- 
ments on the effects of an axially placed magnet on cathode rays. 
They later mounted a circular coil on a Crookes tube between the 
metal cross target and the end of the tube, and found that by 
varying the current through the coil, the shadow of the cross 
upon the end of the tube could be varied in size. They also 
noticed a rotation of the image of the cross on the glass wall, an 
effect which is associated with all magnetic focussing. 

Considering the relatively small size of a Crookes tube in 
comparison to that of a Braun tube, the experiments of Mac- 
Gregor-Morris gave the first demonstration of the image-forming 
properties of a short magnetic coil. It was not until 30 years later 
that the vast implications of these experiments were fully realized. 

Busch, in Germany, published a series of classical papers in 
1926/27 which laid the theoretical foundation for the field of 
electron optics. He proved mathematically that a non-uniform 
magnetic field, in the direction of the tube axis, produced by a 
short solenoid also has the properties of a lens, and that the rela- 
tion between focal length, magnetic field strength and electron 
velocity fitted well known optical formulae. 

The action of a short magnetic lens is illustrated diagram- 
matically in Fig. 86. A divergent electron beam originating at 
P on the electron optical axis of the system enters the non-uniform 
magnetic field produced by the short solenoid, S, and is thereby 
brought to a focus at P'. The short solenoid thus fully satisfies 
the optical conditions for the formation of an image of an object. 
A point in the object is reproduced as a point in the image. In 
optics, the object must be the source of light beams; in electron 
optics, the object at P must be the source of electron beams. 
In optics, the refractive medium is a glass lens; in electron optics, 
the refractive medium is a magnetic (or electric) field. In optics, 
the image is formed on a screen, a photographic plate, or on the 



retina of the eye of the observer; in electron optics, the image is 
formed on a fluorescent screen or on a photographic plate. 

The equivalence of the action of a magnetic lens on an elec- 
tron beam to that of a convergent glass lens on a light beam is 
further illustrated in Fig. 87, where the two corresponding sys- 
tems of lenses are shown side by side. The focal points, F and F x , 
on each side of the lens are also indicated. The distances along 
the axis from the center of the lens to the object P, the image P', 
the first and second focus F and Fi respectively, are related by a 
mathematical equation which applies in both cases. 

— — — »— ^ j ^*"^ 

/ _ — — ^._ \ 

,' \ \ 

Fig. 86. The focussincr action of the field of a short solenoid. 

The transformation of an object point into an image point 
has been demonstrated in the previous examples for cases in which 
both these points were situated on the axis of the system. An 
object of physical dimensions must of necessity extend beyond 
the axis. Just as in light optics, the rule applies in electron optics, 
namely, that an image of objects can be formed as long as the 
image-forming rays do not diverge too far from the axis. Rays 
which travel near the axis are called paraxial rays. The means 
employed to insure that this condition is fulfilled are very similar 
for light rays and electron rays: it is the aperture stop familiar 
to everyone from the photographic camera. Fig. 88 illustrates 



the function of this device for limiting the beam angle. Thus in 
Fig. 88a, the light emitted by the object O at rather large angles 
with respect to the axis is stopped by the aperture disc A, so that 
it cannot reach the lens. It has been mentioned in an earlier 
chapter that such wide-angle rays would cause severe distortions 
of the image if they were permitted to penetrate the marginal 

... P' 


Fig. 87. Comparison of a convergent light (glass) lens and a convergent electron 

(magnetic field) lens. 

regions of the lens. This state of affairs also prevails in electron- 
optical lens systems. Fig. 88b shows the electron rays originating 
at an oxide cathode, C, which is shielded by a guard ring, S. The 
anode, A, which accelerates the electrons, serves at the same time 
as an aperture stop and insures that only paraxial electrons reach 
the magnetic lens, L. 




• ••• 

(b) c 


i 1 



! * 


__!____ ■ ' 






Fig. 88. Comparison of the use of aperture stops in light and electron lens systems. 

The Iron-Shielded Magnetic Lens 

When speaking of lenses in ordinary optical instruments a 
distinction is made between "thin lenses" and "thick lenses." 
These terms are used not only in their everyday meaning of thin 
and thick, but also to indicate that a different method of ray trac- 
ing or calculation must be followed in the two cases in order to 
determine the properties of the single and compound lens, respec- 
tively. A lens is called thin when its thickness along the axis is 
a small fraction of its diameter. This means, in turn, that the light 
ray travels only for a short fraction of its entire path through the 
medium of higher refractive index. 

In the light of this terminology the introduction of Busch's 
short solenoid instead of the long solenoid was a step toward a 
thinner lens. But in optical language this Busch lens is still thick, 



since the magnetic field strength along its axis is a considerable 
fraction of the maximum intensity at the center of the coil for 
quite a distance from the center of the coil on each side. Fig. 89 
illustrates this condition. 

It was recognized by Gabor in 1927 that the magnetic field 
could be concentrated along a shorter distance along the axis of 
the coil if the latter was encased in iron except for the inner wall 

ooooeoooo 9PJP222. 
oooeoooe eeeeo 
o oeoeeeeeoeoe 

r rVr r ? r _"_"-"-'- r~: 

" ""' u- 7 

' ' "* *"<*• ** <• 

:- s z "i ~-~- ~- r z H^: 


/ / ,' t 

t > ' 

e oooooooooooo 
e o oooooooooo 

N \ \ \ x 
1 \ l 

Fig. 89. Busch's original short solenoid magnetic lens. The graph shows the dis- 
tribution of magnetic field strength along the axis. 

of the coil where a non-magnetic metal is used. This is shown in 
Fig. 90; the cross-shading here indicates the winding, and the 
solid black frame the soft-iron enclosure. It is evident that the 
stray magnetic field is reduced considerably, since the magnetic 
field lines prefer to run through the iron casing. 

Knoll and Ruska in 1931 carried this restricting influence of 
the iron shield to the extreme and utilized the stray field at a 



narrow gap as the active field of the magnetic lens. Fig. 91a 
shows this type of coil in its earlier form and the field pattern 
obtained with iron filings. Fig. 91b gives the field distribution 
along the axis of the coil. 



n ( I *TTT< " I 

Fig. 90. 

Solenoid encased in soft iron 
to concentrate the field at the 
center. The graph shows the 
distribution of magnetic field 
strength along the axis. 

b. H 

It was found that a reduction of the coil current by 30 per 
cent results from this type of construction, that is, a required field 
strength can be obtained with about two-thirds of the coil current 
that would be necessary in an unshielded coil. The distribution 
shown in Fig. 91b was obtained when a current of 100 ma. was 
sent through the coil shown in Fig. 91a. The winding of this coil 
consisted of 6500 turns of 0.2-mm diameter enameled wire, and 
it was operated from a 220-volt source. The focal length of this 
coil was of the order of a few centimeters. As a general rule it is 
stated that the minimum focal length of such a coil as shown in 
this figure is from one-third to one-quarter of the inside diameter 
of the iron shield. 

The focal length of a magnetic lens is controlled by the inten- 
sity of the current which flows through the winding; it can thus 
be varied within a considerable range. The minimum value of 



the focal length, which should be as small as possible if large 
magnifications are to be obtained, is determined by the geometry 
of the coil. Thus, when the physical dimensions of a magnetic 
lens are reduced to one-half their original size and the coil current 
is kept constant, then the power of the lens will be doubled or the 
focal length reduced by half. As the magnetic field intensity, H, 
depends on the number of ampere turns (n X /, number of turns 


\UmjnJ; ,>, 

10 cm 

i / / 


Fig. 91. 

Magnetic lens of 
Knoll and Ruska (1931). 
The first use of narrow 
gap in soft iron casing. 


times current in amperes) it is necessary to keep (n X /) constant 
for the smaller coil. Thus reduction of the coil to one-half its size 
means that the current / will flow through a wire which has only 
one-quarter the cross-sectional area, since in order to allocate the 
same number of turns in half the space a reduction of the wire 
diameter by one-half is also necessary; this will lead to overheating 
of the wire if the reduction is carried too far. 



Ruska overcame this difficulty (1934) by the introduction of 
pole pieces which are inserted in the threaded inner wall of the 
coil and in turn contain an annular gap made of non-magnetic 
material, such as brass, where the stray magnetic field protrudes 
and provides a concentrated magnetic field near the axis of the 
coil. The physical size of the coil itself can thus be kept con- 
veniently large so as to prevent overheating of the winding; but 
the field is concentrated into a very small zone by the use of 
accurately machined pole pieces of very small dimensions. A 
magnetic lens of this type is shown in Fig. 92 in a schematic dia- 
gram which does not contain the details to be found in a modern 









Fig. 92. Modern magnetic lens as used in magnetic electron microscope. 

magnetic lens. The correct shaping of the pole pieces is a matter 
of great importance and has a decided effect on the quality of 
the image. 



Courtesy R.C.A. Laboratories 
Views of a sample of aluminum oxide taken with electrons with velocities 
corresponding to different fields (kilovolts) . 


Courtesy University of Michig 

Evaporated zinc on collodion film (X55,0€0) 

Chapter 15 


Historical Background 

When we reflect for a moment on the contents of the previous 
chapter, we realize that the magnetic lens was developed experi- 
mentally without invoking in any way the wave concept of the 
moving electron. The theoretical foundation outlined by Busch 
indeed deals with the electrodynamics of the charged particle and 
not with its wave nature. 

In the development of ordinary light optics also we know that 
lenses were used to form images of objects long before the dual 
nature of light was established. Even today, the designer of 
optical instruments leans heavily on the corpuscular concept in 
the sense in which it was associated with Newton, but has also to 
keep in mind that the fact that waves are involved limits his simple 
geometrical theory. He is thus very old-fashioned and very 
modern at the same time. When he wishes to determine the 
path that a light beam will follow through a number of lenses, he 
proceeds to draw straight lines according to geometrical rules. 
We have used this technique in Chapter 2. 

The fielchof application where this method gives satisfactory 
results is called geometrical optics. It covers the effects of reflec- 
tion and refraction and lends itself particularly to ray tracing 
on mirrors and through prisms and lenses. The effects which 
require the wave concept for their explanation form the domain 
of physical optics. Here we find the treatment of diffraction, 
interference, and polarization. Since we realize that the corpus- 
cular and the wave concepts cannot be segregated in principle 
when dealing with the nature of light, some one may object to 
this classification of geometrical versus physical optics. After all, 



it is the same light which produces the effects in both domains. 
Nevertheless, this terminology has been carried through the text- 
books and may have its use at times. 

We should then, likewise, speak of geometrical electron optics 
in cases when the corpuscular concept of the electron is upper- 
most in our minds, and of physical electron optics when we stress 
the wave nature of the electron, while being fully aware of the 
dual nature of the electron at all times. This practice has indeed 
been followed. 

When we trace the trajectory of an electron beam through 
an electrostatic potential field we use a set of geometrical rules — 
different of course from that used in optics — and obtain a first 
approximation to the actual path. For this reason, the treatment 
of electron lenses and the associated electron paths forms the sub- 
ject of geometrical electron optics. Davisson and Germer's ex- 
periments on the diffraction of electrons in a crystal lattice and 
the discussion of the resolving power of the electron microscope, 
where the wave nature of the electron plays the decisive role 
belong to the field of physical electron optics. At times it may 
be hard to decide which treatment will solve a certain problem 
the more easily. 

We may now state that geometrical electron optics was devel- 
oped as a result of — and not before — the full understanding of 
the dual nature of the electron. This need not have been so. 

Long before the electron was discovered, Sir William Rowan 
Hamilton had clearly shown (in 1830) that a striking analogy 
exists between the law that governs the movement of a material 
corpuscle through a given field of force and the law which gov- 
erns the trajectory of a light beam through a series of media 
of different refractive indices. 

In the first case, the mass will move from a point Pi to a point 
P 2 under the influence of the existing mechanical forces in such 
a manner that the sum of all the products of the momentum, mv, 
multiplied by the length of a small element of the path, ds, will 
give the smallest possible value when added up over the total 
length of the actual path from P x to P 2 . Readers who have been 
exposed to a mathematical treatment of mechanics will recognize 


in this rule the famous principle of Maupertuis, formulated in 
the middle of the eighteenth century, which may be given the 
mathematical form: 


p 2 
vds = a. minimum (Mechanics) 

In the case of light, Fermat had shown, in 1667, that a beam of 
light will follow a path from a point P x to a point P 2 through 
various media of different refractive indices in such a manner that 
it takes the least possible time. This principle can be expressed 
mathematically as: 

nds = 2. minimum (Optics) 


where n is the refractive index of the medium at any point. 

Putting these equations side by side reveals the fundamental 
similarity between light and matter. We sense that the particle 
velocity, v, in mechanical laws takes the place of the refractive 
index, 72, in the optical laws. 

When the electron was discovered and recognized as a particle 
of a definite mass, m, and electron experiments had revealed strik- 
ing similarities to optical experiments (1900), the way was open 
to develop geometrical electron optics on the basis of Hamilton's 
analogy between the basic mechanical and optical principles. By 
the time the dual nature of light was generally accepted by 
physicists, about 1910, there was really no excuse for not suspect- 
ing a like duality of the electron. Evidently, such a sympathetic 
approach to the study of nature, which would have been natural 
to the Greeks, was not cultivated at the time. 

So the years went by; existing theories were carefully refined 
and much experimental work done with the aim of revealing the 
structure of the atom and the role which electrons played in it. 
Many times, when electron beams were caused to interact with 
atoms, whether they were in the form of a gas or in a solid array 
at the surface of a crystal, the observed effects were not in agree- 
ment with the conventional view of the corpuscular nature of 
the electron. 



As we have seen, it was not until 1923 that de Broglie devel- 
oped his electron wave theory and a few years later that Davisson 
and Germer and, quite independently G. P. Thomson, showed 
experimentally that a wave motion must be associated with an 
electron beam. 

In 1926-27, Busch established the theory of the electron lens 
represented by axially symmetric magnetic or electric fields, and 
thus became the founder of geometrical electron optics. In 1931 
Davisson and Calbick performed experiments on the lens proper- 
ties of apertures and published the formula for the focal length of 
a circular and of a slit aperture. Thus the electrostatic electronic 
lens made its appearance. We shall now describe such lenses in 
more detail. 

Electrostatic Lenses for Electron Beams 

If we arrange three metal discs A, B, C, in a vacuum tube and 
assign to them the potentials as shown in Fig. 93a we may plot the 
potential values that exist along the axis O-CX of the system. 
Discs A and B are at zero potential and disc C is at 100 volts. 
The space between A and B is thus at zero potential, if the stray 
fields at the edges are neglected, and the potential plotted in Fig. 
93b along the horizontal axis, Z, remains zero. In the space 
between B and C a potential gradient exists. The field is constant 

P\ B C 



I I I 

I I 
i I i I 

20 40 60 8 °-HOOV 




Fig. 93. Distribution of the electrical field between conducting plates at different 



and has a definite positive value. This is indicated by the equi- 
distant equipotential planes, represented by the thin vertical lines 
between B and C in Fig. 93a, and accordingly a linear rise of the 
potential plot in Fig. 93b along the axis between B and C. 

In Fig. 94a the same electrode arrangement is shown, but the 
disk, B, is provided with an aperture. The field to the right of 
B now bulges through the hole in B into the space to the left of 
B, where previously no field existed. The equipotential planes- 
become curved in the vincity of the aperture and straighten out 
only in front of plate C. The potential values are written on 
the potential areas in small figures. Fig. 94b gives the correspond- 
ing rise of potential along the axis. This plot is no longer a 
straight line throughout, but contains only a short straight section 
in front of C. The slope of the potential plot is small, i.e., the 
field is weak to the left of B, and it increases rapidly to the right 
of B. It is worthy of note that the potential at the center of the 
aperture is greater than the potential of the disc itself. 

Let us now assume (Fig. 95) that disk A is an electron emitter 
at the center. This is readily realized in practice by inserting 


,.-•■-:;;; / 

I I 

Of I ■ 10, 15 £ ' io ( . ''■so/ 4Q Sol 60J 7o( 6 i 



Fig. 94. How the potential distribution is affected by an aperture in an electrode. 



the oxide-coated face of a cathode sleeve at the center; this is 
heated by a flat spiral heater wire back of it. The cathode sleeve, 
generally made of nickel, is connected with the main disc A by 
a thin wire, w, so as to maintain zero potential on both of them. 
The disk A thus extends the zero potential face of the cathode 
without draining any noticeable amount of heat from it. This 

o 100V 

Fig. 95. The influence of an electrode with an aperture near the cathode. 

arrangement is frequently used in cathode-ray tube practice and 
is known as a guard ring. When a current is sent through the 
heater wire, the oxide-coated face, F, will reach a dull red tem- 
perature and electrons will be emitted from it. The paths of 
electrons near the axis are shown brought to a focus, Fi. 

Electrons Emitted with an Initial Velocity 

We have previously assumed that electrons are released at 
the cathode surface with zero velocity. This is true for some 
electrons; others possess an initial velocity even in the absence of 
an electric field in front of the cathode. Furthermore, while most 
of the emitted electrons leave the cathode face at right angles in 
the direction of the Z-axis, others are emitted at an angle with 
respect to the normal. This is indicated by the three small 
arrows in Fig. 95 emanating from the center of F. We may illus- 
trate this space distribution of the electrons emitted from a point, 
P, at the cathode, C, in Fig. 96. Here a circle is drawn in front of 
the cathode face and a number of arrows are inscribed, all of 


which originate at P. The lengths of these arrows indicate the 
relative number of electrons emitted in the direction given by 
the arrows. It becomes clear from this diagram that many elec- 
trons are emitted in a direction normal to the cathode face, but 
that a definite fraction of the total emission occurs sideways. 

It may then be stated that electrons are emitted from a cathode 
with different initial velocities and in different directions. An- 
other noteworthy feature of the movement of electrons in an 
electric field must now be explained before the electron paths in 
Fig. 95 can be fully understood. It was stated earlier that elec- 
trons which start from a position of rest travel in an electric field 
in such a manner as to follow the field lines or, what amounts to 

Fig. 96. 

The number of electrons emitted 
from a point on the cathode with 
velocities in different directions. 

the same thing, so as to penetrate the equipotential planes at right 
angles. This was the case in Fig. 58, where we assumed an emis- 
sion with zero initial velocity. 

We have learned in the meantime that the initial velocities 
may have positive values — up to about 0.2 volt, to be specific — 
and that these initial velocities may be directed sideways with 
respect to the normal to the cathode face. How will this affect 
the electron paths? A little earlier, the rectilinear movement of 
an electron in a uniform electric field from a position at rest was 
compared with a stone that falls to earth in a straight line in the 
gravitational field when released from a position at rest in the 
hand. If we now give the stone an initial velocity that is directed 
at an angle to the previous path, the trajectory of the stone will 
be a parabola, as shown in Fig. 97. The horizontal lines 1, 2, 3, 
4, 5, represent equipotential planes, i.e., points where the work 
that must be done to raise the stone from the ground level to the 



height indicated is constant. We note from Fig. 97 that the stone 
moves at right angles to the equipotential planes from O to A, 
when it starts either from a position at rest or when its initial 
velocity is directed at right angles to the equipotential planes. On 
the other hand, when the initial velocity is directed sideways, the 
path O-B is no longer at right angles to the equipotential planes. 
The stone is then refracted at the respective equipotential planes. 






/ ■ 




i a 



Fig. 97. The path of a falling stone which has been thrown out with a velocity 
in the direction of the arrow. 

The same reasoning applies to an electron which moves in 
a uniform or non-uniform electric field with an initial velocity 
which makes an angle with the field lines. This is illustrated in 
Fig. 98. The electron trajectories in Fig. 95 are now readily 
understood. The electrons which are emitted in the direction 
along the axis travel in a straight line to Fi on plate C, just as the 
axial ray in an optical system reaches the focus in a straight line. 
The focussing action of the potential field is exerted on the "stray 
electrons" which are emitted sideways at angles not too large. 

As has been stated, the focussing action of an aperture system 
upon an electron stream was first recognized and expressed in 
mathematical form by Davisson and Calbick of the Bell Tele- 
phone Research Laboratories (New York) in 1931. These 
workers published the formula for the focal length of an electro- 



Fig. 98. The paths of electrons emitted with a velocity making an angle with 
the direction of the lines of force of the electrical field. This explains directions 
indicated in Fig. 58. 

static electron lens of the aperture type expressed in terms of the 
voltages applied to the electrodes. Briiche and Johannson of the 
Allgemeine Elektrizitats Gesellschaft (A.E.G.) in Berlin pursued 
similar work independently at about the same time. The German 
school was particularly successful in applying these new electron 
tools to the design of practical instruments, one of which was 
the electron microscope. The principles underlying these aper- 
ture lenses, as they are called, deserve closer study. 





Fig. 99. How the path of an electron bends when travelling through a potential field. 

The Refraction of Light and Electrons 

In Fig. 99 the potential distribution between two plates, A 
and B, is shown for the case where plate A is kept at a potential 
of +500 volts and B at a potential of +1500 volts. The areas of 
equipotential will be distributed symmetrically as marked. 


If an electron beam with a velocity equivalent to 1,000 volts 
enters along the axis O-d from the left, it will follow a curved 
path as shown and strike the plate B at P, if the potential of B is 
high enough, or the plate sufficiently long. The layers of increas- 
ing potential thus act upon the electron beam in much the same 
way as the layers of differing refractive index act upon the light 
beam. Fig. 2 is repeated here in order to bring out the analogy 
that exists between the refraction of light in the layers of the 
atmosphere of different densities and this electron refraction. 

Apparen t position 
of sun 

Real position 
of sun 

Fig. 2 repeated to show clearly the analogy between the refraction of light and the 
bending of the path of electrons. 

The refraction of a light beam when it passes from one 
medium, such as air, into a denser medium, such as water, is 
accurately described by Snell's law to which reference was 
briefly made on page 45. Fig. 100 describes the experimental 
conditions which serve to demonstrate this fundamental law of 
optics. The line AA' with its shading represents the boundary 
between two different optical media which are characterized by 
the refractive index n x above the boundary and n 2 below it. Thus 
AA' may represent a water surface with air above and water 
below the boundary. If a narrow beam of light produced by the 


lamp L, behind a screen, S, is directed toward the boundary, AA', 
where it impinges at O, it will continue its path in the water in 
the direction OP. The vertical line, NN', represents the normal to 
the surface, AA', at O. The angle a ± is then called the "angle of 
incidence" and the angle a 2 the "angle of refraction." These 
two angles can readily be measured. It is found that the light 
beams above and below the surface and the normal at the point 
of incidence always lie in one plane, which in Fig. 100 is the 



.>>))>) >>>>)>> 




Fig. 100. (a) The law of refraction. (b) The definition of sin «. 

For any angle a, if we draw any right-angled triangle containing a as one angle, 
the ratio of the side opposite the angle (AB) to the hypotenuse (CA) is called the 
sine of the angle a and is written sin «. 

plane of the paper on which the diagram is drawn. When the 
light source with the screen is moved in the plane of the paper on 
a half circle, with O as center, the angle of incidence, a u will 
take on different values, and the angle of refraction a 2 will also 
change. It was found by Snell that, irrespective of the position 
of the light source above the boundary, the ratio of the sine of 
the angle of incidence to the sine of the angle of refraction is 
identical for all positions of the light source, as long as the medium 

above or below the boundary is not changed. This ratio — : 

J a sin a 2 

is called the index of refraction of the water. If n indicates the 

index, n = 

sin ot! 
sin a 2 

Strictly this definition is true only when the 

upper space is a vacuum, but air differs very little from a vacuum 


in its effect on the bending of the ray at O. In the case of air and 
water as the two media, the ratio will have the constant value 
1.33. If the water is replaced by glycerin, the ratio becomes 
1.47. When the experiment in Fig. 100 is performed between 
two liquids or two different kinds of glass, with indices n x and n 2 , 

the ratio L is equal to the ratio ~. This general form of 

sin a 2 n t 

Snell's law governing the refraction of light passing from a 

medium with index n t to a medium with index n 2 may be written: 

n-i sin <xi — n 2 sin a 2 

According to Sir William Rowan Hamilton's conception of 
the identity of the optical laws describing the passage of a light 
ray through a series of refractive media with the mechanical laws 
describing the motion of a small mass through a potential field, 
we are led to expect the validity of a modified form of Snell's 
law applying to electrons moving through a field of electrical 
potentials. This belief is based on our conviction that an electron 
is a particle of a small but uniquely defined mass. The com- 
parison of the two basic principles which govern the path of a 
light beam and that of a moving mass, respectively, (p. 177) 
indicated that the electron velocity v should take the place of 
the refractive index n in the optical law, so that we would obtain 
by analogy a law governing electron refraction thus: 

Vi sin a 1 = v 2 sin a 2 (1) 

It remains to be shown that this is correct. 

Fig. 101 is an enlarged view of a small portion of Fig. 99, 
where an electron beam crosses the equipotential plane marked 
1200V. It might apply to any one such potential plane cross- 
ing. In order to make Fig. 101 a more complete counterpart of 
Fig. 100 and to be intelligible without reference to Fig. 99, the 
light source, L, in Fig. 100 is replaced by an electron source or a 
small electron gun. The shield, S, in front of the cathode, K, is the 
accelerating anode at potential V. The boundary, AA', separates 
two regions of different potential, V x above and V 2 below the 
boundary and it is assumed that V 2 is larger than V x , so that the 


electrons are accelerated when crossing the boundary. Thus 
there exists a potential gradient at the boundary in the direction 
NN' normal to the surface. This gradient represents a force 
which will act on the electrons when they reach the point O. 

In order to find out how this force affects the trajectory of 
the electrons we resort to an artifice which we used once before 
(p. 152) in a similar problem. Since the electron beam travels 
in a field of uniform potential Vi above the boundary, its trajec- 
tory is a straight line and the velocity v 1 of the electrons is con- 
stant along this path from the aperture S to O. This velocity is 
a directed quantity and can thus be represented by a vector XO, 
where the origin of the vector is an arbitrary reference point 
along the electron beam. We split up this vector v x into two 

Fig. 101. Refraction of electron beam. 

components at right angles to each other where v x ' is the resolved 
part in a vertical direction and v x " the part along the boundary 
AA'. We then consider what happens when the electron at X 
has reached the point O, where the force due to the potential 
gradient acts upon it. Since this force lies in the direction of the 
velocity component v t \ the velocity in the direction ON' will be 
increased while the velocity component v^" will be unaffected 
since no force acts in its direction. The two vector components 
Vi and v 2 " (which equals v±) add up to the vector v 2 , which 
gives the new direction and velocity of the electrons in the region 
below the boundary where the potential is V 2 . If we now relate 
the velocity vectors to the angle of incidence, a 1? and the angle of 


refraction, ct 2 , we obtain the following expressions by putting in 
the values of sin a x and sin a 2 in the two vector triangles in 
Fig. 101: 

sm(x 1 = v 1 "/v 1 (2) 

sin <x 2 = v 2 "/v 2 (3) 

By dividing equation (3) into equation (2) we get 

sinai v 1 "v 2 //lN 

■ = (4) 

sin a 2 v 2 "V\ 

Since v'x" equals v 2 " they cancel out, so that 

sin <zi v 2 

sin <x 2 v-l 


vi sin (x^Vo sin <x 2 (5) 

Equation (5) is identical with equation (1), which we deduced 
by analogy. It is thus established that the electron velocity v 
represents the electron optical refractive index. As we have 
shown (p. 134) the electron velocity is determined by the 
potential difference through which the electron has been accel- 
erated; it is related to V by the equation 

v = K^Y (6) 

where K is a numerical constant equal to 5.93 X10 7 when V is 
expressed in volts and v in cms. per sec. We can substitute equa- 
tion (6) in equation (5) and obtain 

sin a'i \/V 2 

sin a 2 yFi 

The fundamental law of electron optics is often quoted in this 

We shall now proceed to give in the following chapters a 
description of the structural features of the electrostatic and mag- 
netic electron microscopes as far as it is essential to the under- 
standing of their operation. Many details must of necessity be 
omitted in a general text of this type. 


Courtesy American Cyanamid Co. 

Replica of surface scales on wool fiber, X 9,000 (3,000 X electronic, 3X optical). 



Courtesy University of Michigan 

Replica of grating with line interval of 1.78/x. 

Courtesy University of Michigan 

Same as above, illustrating method of calibration. The distance between the lines 
is known and magnification is changed by changing the current in lens coil. 

Chapter 16 


Object and Image 

In order to form images of objects with the aid of electron 
beams, one essential condition must be satisfied. Electrons which 
leave a point on the object at small angles with respect to the 
normal to the object plane must be reunited at a point in the 
image plane. In order to bring this about the object and image 
planes are parallel to each other and separated by a given distance 
along the electron-optical axis of the system. Object points in 
the object plane are confined to the close proximity of the axis, 
whereas image points in the image plane are located within the 
circle of sharp image formation which may have a diameter of 
several inches, depending on the magnification of the system. 
These statements are illustrated in Fig. 102 in a simplified manner. 
Five representative points A, B, C, D, E are chosen in the object 
plane (O, P) and form the apex of five cones of divergent elec- 
tron rays. In the plane indicated by L an electron lens, or a 
combination of several such lenses, which in effect can be replaced 
by an ideal thin lens, is located. The function of this lens is two- 
fold. The cone of electron rays which diverges from the object 
point must be inverted so as to converge to a point in the image 
plane. The lens plane thus becomes the base of a series of double 
cones. In order to bring about this result the forces acting upon 
the electrons in any one of the various beams must be directly 
proportional to the radial distance of an electron from the axis 
of any one beam to which it belongs. In this manner, the elec- 
trons which tend to diverge the most undergo the greatest con- 
stricting force. At the same time, the electron lens will impart 
to the beams a deflection which will make the distances between 




the image points, A'B'C'D'E', respectively, greater than the corre- 
sponding distances between the points, A B C D E, in the object 
plane. Thus a magnification of the object results which is 
given by the ratio of the distance of an image point from the 
optical axis, CC, to the distance of the corresponding object 
point from the axis. This ratio M, the magnification, must be 
constant for all image points in order to obtain a linear repre- 
sentation of the object in the image plane. 

^ r rrrrrrmTm7 lllllll nn ll u l h)7Tr. 

^^nHm/ W77m77T77mTTT7Trrrrrrrr^ 

Fig. 102. Object points and corresponding image points produced with a negative 

electron lens. 

Figs. 102 and 103 illustrate how the electron beams may either 
spread out from the lens plane or cross each other in an inter- 
mediate plane, F, near the axis. (It may be remarked here, that 
this so-called "cross-over" is of particular interest in the theory of 
cathode ray tubes since it represents a virtual cathode of high 
current density. A limiting aperture is placed in this plane, F, 
and an image is formed of it on the fluorescent screen with as low 
a magnification as possible so as to obtain an intense and very small 
pin point of light on the screen.) 

An Electrostatic Negative Lens 

In the first case (Fig. 102) the lens action would be equivalent 
to a negative lens in optics (see Fig. 12b) and could be brought 



Fig. 103. Object points and corresponding image points produced with a positive 

electron lens. 

about by an aperture lens, A, in front of a cathode, K, followed 
by a field free space to the right of A (Fig. 104). This arrange- 
ment represents the only negative electron lens known. All 
others are positive in their action (see Fig. 12a). Since the lens 
field in Fig. 104 extends to the cathode, which now becomes 
our object in electron optics, we have to classify this arrange- 
ment as an immersion lens. Unfortunately this system has no 
practical value because the voltage that can be applied in close 


111 ' 
ii 1 ! 


\ i ii 
■ i 1 1 

\ 5 


i i i i 
1 i i i 

i i i i 

' \ i • 


• 1 J i ! 
i j i I 

! i i i 
I i ' i 
i , i i 
i • 

• i i i 

1 \\\ 


i i i| 

o +• 

Fig. 104. The theoretical arrangement for a negative electron lens. 



proximity to the cathode is limited to values of the order of 50 
volts, and electrons of such low velocity barely excite fluores- 
cence on a luminescent screen. We find then, in most electron 
image-forming systems, that the beams cross the optical axis and 
that the position of the image is inverted with respect to the 
object, as shown in Fig. 103. 

Electrostatic Positive Lenses: The Three-Electrode Lens 

The first practical electron lens system which lent itself to 
the construction of an electrostatic electron microscope was de- 

S A 




Fig. 105. Briiche and Johannson electrostatic electron microscope. 

Diameter C = 
Aperture G= 
Aperture A= 

c = 


5 mm. 
1.2 mm. 
1.0 mm. 
0.5 mm. 

1.0 mm.* 

Dimension d is distance from G to A, not from K to A as shown in the figure. 

L=240 mm. 

V G =140 volts 
V A =750 volts 
V s =750 volts 

scribed by Briiche and Johannson in 1932. It is called "the 
immersion objective lens." We shall here briefly describe the 
structural features of this lens without going into a discussion of 
its theory, which is rather involved. For purposes of brevity 
we shall use the letters I.O. as an abbreviation for immersion 

In its original form, the I.O. consisted of two aperture disks 
located in front of the cathode, as shown in Fig. 105. The indi- 
rectly heated cathode, K, of the oxide-coated variety, is sur- 
rounded by a guard ring at cathode potential as we have described 


on other occasions in this text (see Fig. 95). In front of it, at a 
distance, c, from the cathode surface, a first disk of molybdenum 
sheet, 0.2 mm thick, is mounted with aperture G, 1.2 mm in 
diameter. In conformity with radio-tube practice, this disk 
element is called the grid, G. At a distance, d = l mm, from G, 
a second molybdenum disk, A, with an aperture 1 mm in diam- 
eter, is mounted; this serves as anode. The fluorescent screen, 
S, which is maintained at anode potential, is mounted at a distance, 
L = 240 mm, from the cathode surface. It goes without saying 
that the various electrodes must be aligned most accurately with 

O o 


Mi i i l .i. rLi J L!J* \ 7 T.LL m' 

i r rrm nniii i ini" 

Fig. 106. Arrangement of part of the electron microscope on an optical bench 
built inside a vacuum tube. 

respect to the axis and with respect to one another, since the 
slightest deviation from axial symmetry will produce image dis- 
tortions. The aperture edges must not only be perfectly round 
but also free from burrs and carefully polished. In practice, this 
accuracy of the mount is achieved by the use of ceramic spacers 
between electrodes or by jig mounting in glass beads which are 
softened during assembly so that the electrode supports can set in 
the glass while the proper distances are maintained by the jig. In 
an experimental setup such as that of Johannson, the elements 
were mounted on an optical bench along which they could be 
moved in vacuo by means of external magnets. Provision was 
also made for varying the distance, c, between cathode and grid 
aperture, G, by mounting the cathode on a system of bellows. 
Fig. 106 shows the elements on the optical bench support, Fig. 
107 shows the cathode bellows, and Fig. 108 a cathode image 



obtained with the LO. by C. E. Hall at Toronto in 1936. By 
adjusting the voltages on the electrodes C, G, A, Briiche and 
Johannson found that a sharp image was obtained for Fg=140 
volts and Va = 750 volts corresponding to the dimensions given 
in the legend of Fig. 105. 

f ^ M 

1 1 

b — ^~ 






Fig. 107. Arrangement of bellows for adjustment of the position of the cathode 

in a three-electrode lens. 

Fig. 108. Image of cathode with electrostatic lens. 

The Magnification of a Three-Electrode Lens 

Fig. 109 shows how the grid potential Vg must be adjusted 
in order to maintain a sharp image when c is varied. Fig. 110 
gives the resulting effect on the magnification, M. 

It turns out that, for any given value of c, a sharp image is 
obtained with a definite value of the ratio Fg/Fa, and the mag- 
nification, M, depends only on this ratio, irrespective of the mag- 
nitudes of Vq and Fa, which satisfy this ratio. This is simply 
an experimental verification of the general law for electrostatic 
electron lenses which we derived above; the power of such a lens 
depends always on a potential ratio. In Fig. Ill the magnification, 
M, the diameter of the sharply reproduced image B , the diameter 


200 r 

Fig. 109. Graph showing the value of the grid potential, V G , necessary to give a 
sharp image in the three-electrode lens as the distance, c, between the cathode 
and the grid is changed. 

of the sharply reproduced region on the cathode A = B /M and 
the distance, c, are plotted against the ratio Vg/Vk as abscissa. 

If one aims to reproduce sharply as large an area of the cath- 
ode as possible, the I.O. must be operated at a value 7g/7a = 0.1, 
for which A is a maximum in Fig. Ill, and c must be made equal 
to 1.0 mm. The magnification then is about fifty-fold. We also 
learn from the characteristic curve that M reaches values as high 
as 200 if we are satisfied to reproduce smaller areas of the cathode. 
The cathode must then be very close to the grid aperture and the 
latter will assume negative potentials according to Fig. 109. This 
is analogous to light microscope technique, where large magni- 






Fig. 110. 

Variation of the magnifica- 
tion, M, of a three-electrode 
lens with the distance, c, be- 
tween cathode and grid. 

2 3 

C mm. 



Fig. 111. 

How (1) the magnification (M), 
(2) the diameter of the sharp image 
(Bo), (3) the diameter of the region 
on the cathode (A =B /M) which 
is sharply reproduced in the image, 
and (4) the distance between cathode 
and grid (c), severally vary with the 
ratio of grid potential (V G ) to anode 
potential (V A ), for the three-elec- 
trode lens. Bo, A , and C are ex- 
pressed in millimeters. 



fications are obtained with an objective lens of very short focus 
which demands that the object must be placed almost in contact 
with the lens. 

The Four-Electrode Lens and its Magnification 

A modified I.O. of superior quality was described by Johann- 
son in 1934. It differs from the system just described by having 
an additional electrode which makes it a four-electrode system. 
Fig. 112 gives the diagrammatic view. The performance of the 
four-electrode I.O. is superior to that of the three-electrode sys- 
tem by about 50 per cent, as measured by a figure of merit B/L; 
B is the diameter of the sharply reproduced image. It makes pos- 
sible not only a larger useful magnification, by 25 per cent, but 


also permits of the variation of M over a wider range. The focal 
length is correspondingly reduced to the order of 1 mm. 

Much larger magnification can be obtained with the electro- 
static electron microscope if all the dimensions of the I.O. are 
reduced by a constant factor. M = X 4,000 has thus been reported 
by Johannson. Since the size of the aperture must also be corre- 
spondingly reduced we would obtain aperture diameters of the 
order of 0.1 mm (4 mils) when using a reduction factor 10 to 

Fig. 112. Johannson's four-electrode electrostatic electron microscope. 

D c = 5 mm. 

Dgi= 1.1 mm. 

Dg2= 1.0 mm. 

Da= 1.0 mm. 

c= 0.05-1.5 mm. 
di= 0.48 mm. 
d 2 = 0.72 mm. 
L= 240 mm. 

get M= X 2,500. This will necessarily weaken the intensity 
of the beam and reduce the brightness of the image obtained on 
the screen. For this reason the instrument is generally operated 
to give a magnification of the order of a few hundred times. 

The Resolving Power of the Electrostatic Lens 

When we come to consider the resolving power of the I.O., 
we must remember that it depends on the electron wave length 
in the object plane. The initial velocities of electrons emitted 
from an oxide cathode are of the order of 0.1 volt. Introducing 
this value into de Broglie's formula for the electron wave length 
(see p. 123). 


we obtain: 


— — XlO 8 cm 

— — XlO" 8 cm = 38.92xl0- 8 cm 


This value for I must then be substituted in the formula given 
on page 72 so that we obtain for the distance, d, of two object 
points which can just be separated in the image (0.61 replaces %) : 

, 0.61A 0.61X38.92X10- 8 

d=-^r^r- = ^r^ cm 

N.A. N.A. 


XlO- 8 cm 


The numerical aperture (N.A. = n sin A), as defined on page 
72, is much smaller in electron-optical microscopes than it is in 
ordinary optical microscopes where values N.A. = 1.5 can be 
realized with an oil-immersion objective. 

In electron-optical systems severe limitations are imposed on 
the size of the aperture diameter in order to keep lens aberrations 
to a minimum. These aberrations are caused by rays divergent 
from the optical axis beyond a permissible amount, as we have 
learned above. To prevent such excessive divergence is the pur- 
pose of a very small aperture diameter which leads to a reduced 
value of sin A. We have also found in Chapter 15 (page 188), 
that the electron velocity corresponds to the refractive index n 
in optical systems. Since we are dealing with even very slow 
electrons in the I.O. this value becomes very small, so that n sin A 
is smaller by three orders of magnitude in comparison to values 
common in ordinary optics; i.e., N.A. = 0.001. If we introduce 
this value in the last equation we obtain 

23 7 
d=—~—x lO- 8 = 23.7X1000 XlO- 8 cm 

= 23,700 A.U. 

Actually, Briiche and Knecht determined experimentally that 
a resolution of 15,000 A.U. had been reached with their instru- 
ment (1934). 


There are other factors which operate against obtaining high 
resolving power with the I.O. when it is used for the investiga- 
tion of self -emitting objects, in particular of oxide-coated cath- 
odes. The granular structure of the cathode surface will distort 
the first refracting surface of the electron lens, i.e., the equipo- 
tential area which matches the surface of the cathode. A local 
negative space charge in front of the cathode produced by par- 
ticularly active spots on the cathode may act in a similar manner. 
For refined investigations, it is then necessary to provide as 
smooth a surface as possible and to operate it at a relatively low 
temperature, so that space charge distortions are avoided. 

In spite of these restrictions, the service of the electrostatic 
electron microscope in manifold researches has been invaluable. 
We shall give a brief survey in the following chapter of the vari- 
ous applications to which it has lent itself. 

The Possibilities of Using Bright Emitter Cathodes 

We have thus far considered only the case of an oxide-coated 
cathode being used as an electron source in conjunction with the 
LO. Nothing stands in the way of using a bright emitter such 
as a tungsten or molybdenum ribbon as the object. Due to the 
elevated temperature at which these materials have to be operated 
in order to yield emission currents comparable to those obtained 
from dull emitters, a certain amount of direct light will reach 
the central area of the image and produce a background illumina- 
tion. This need not be serious since the apertures are small in 
any case. The metallic surfaces of these cathodes can be highly 
polished so that field distortions are greatly reduced at the surface. 
In addition to this, higher voltages may safely be applied so that 
the resolving power will be increased considerably. 

Photoelectric Surfaces as Cathodes and other Variations 

Photoelectric surfaces can also be used as cathodes in the I.O. 
Provision must then be made that a light beam reaches the photo- 
cathode from the side, or from the rear when a transparent photo- 
cathode is employed. This latter principle is the basis of 
Zworykin's image tube which has found many useful applications. 


Since the emission yield from photo-cathodes is comparatively 
small, it is necessary to apply voltages of the order of 10,000 volts 
to the lens anode in order to obtain images of sufficient brightness 
on the luminescent screen. 

The cathode may be the source of secondary electrons pro- 
duced by the impact of primary electrons either under angle from 
the side or from the rear in the case of a thin metal film. 

Finally, high-speed electrons may be shot through a thin film, 
a technique which was explored experimentally by Behne in 
1936. He was thus able to obtain images of aluminum foil 0.6^ 
thick with a magnification nearly X 100, when 500-volt electrons 
were shot onto the foil from the rear and then accelerated by 
3,000 volts and more on the anode of a three-element, Johannson 
immersion objective. 

Mahl, in particular, has published a series of papers since the 
outbreak of the present war on the high-voltage electrostatic 
electron microscope, which, unfortunately, are not accessible to 
the authors of this book at the present time. Consequently we 
must confine our remarks to information gathered from reviews. 
It is stated that a resolving power of from 80 to 100 A.U. has been 
obtained; this is a vast improvement over the 15,000 A.U. reported 
by Briiche and Knecht in 1934. It represents an improvement 
over the greatest possible resolving power of the optical immer- 
sion microscope, using ultraviolet light (2000 A.U.) by a factor 
of twenty. 

The G. E. Electrostatic Electron Microscope 

During the early part of 1943 a simplified type of electrostatic 
electron microscope was described by Bachman and Ramo, of the 
Electronics Laboratory of the General Electric Company at 
Schenectady (see plate p. 205). This is the first commercial 
design of this type of electron microscope on this continent and 
contains many interesting features. It is comparatively small in 
size, simple in construction and operation, and its resolving power 
is about 10 times better than that of the light microscope, in other 
words, 200 A.U. 

Fig. 113 is a cross-sectional diagram of the General Electric 


Fig. 113. 

Cross-sectional diagram of 
the small G. E. electrostatic 
electron microscope. 

Vacuum c6sm6er 



instrument. The lower portion of the cabinet contains the pump 
unit and the upper portion the power supply and the microscope 
proper. A work table with drawer protrudes from the middle 
and the whole unit is mounted on casters so that it can be moved 
about easily. It is operated on 110 volts. Two minutes after 
insertion of the specimen into the vacuum chamber, observations 
can be made on the screen in front of the observer. 

Fig. 114a gives a schematic diagram of the electron-optical 
components of the electrostatic microscope under discussion. 
A tungsten filament, F, of hairpin shape serves as the source of 
electrons which originate at the tip of the filament where the 
temperature of the filament is highest. We are thus dealing with 
a filamentary cathode in distinction to indirectly heated plane 
oxide-coated cathodes which were used in the earlier German 
models of this type. The filament is surrounded by a cylindrical 
shell, K, to which one side of the filament is connected. The 
position of the filament tip with respect to the aperture of the 
surrounding shell can be adjusted by means of side screws and the 
shell in turn can be lined up accurately with the microscope axis. 
The filament and shell are maintained at a negative potential 
which can be varied smoothly by a variac mounted on the front 
panel from a low value up to about 35,000 volts. The filament 
current is likewise controlled by a variac similarly mounted on 





Fig. 114. 

(a) is a schematic diagram 
of the optical components of 
the G. E. electrostatic electron 
microscope, (b) shows the 
potential variation along the 
axis of the microscope. 


the front control panel. Leaving the cathode aperture and travel- 
ing along the optical axis, the electrons are accelerated to a high 
velocity equivalent to the voltage applied to the anode A. They 
maintain this velocity in the spaces denoted in Fig. 114a by 1, 2, 3, 
but slow down momentarily in the intermediate lens regions Li 
and L 2 . The potential plot in Fig. 114b indicates the potential 
variation along the axis brought about by the electrode con- 
figuration of Fig. 114a. 

The representation is similar to that used earlier in Fig. 94 
(p. 136) . As a matter of fact, the latter might well represent the 
first part of our potential plot in Fig. 114b, where it rises steeply 
from the origin. After reaching the full value of V in space 1, 
the potential drops to nearly half, as indicated by Vi in the lens 
region Li and L 2 which are separated by the space 2 and followed 
by 3 where the electrons again reach the full value V . The po- 
tential troughs at L x and L 2 represent the main electron lenses 
which form an image of the object O on the screen S. Since the 
potential rises to a constant value V on either side of the lens 
plane, this type of lens is called a unipotential lens. 

It is worthy of note that the electron gun proper, comprising 
F, K, A, and the two lenses Li and L2, are operated from a single 
voltage supply, V . This makes for great simplicity of operation. 


Furthermore, since the focussing action of the electrostatic lenses 
depends on a voltage ratio V /Vi rather than on any single volt- 
age, voltage fluctuations will not disturb this ratio and the focal 
length of the lenses is thus kept constant regardless of such fluc- 
tuations. This eliminates the need for elaborate voltage regu- 
lators in the power supply unit. 

The image produced on the fluorescent screen, S, which may 
be replaced by a photographic plate, is further magnified about 
seven times by a light optical system. An electronic image on 
the screen giving a magnification of X 1000 may thus be further 
enlarged to a total magnification of X7000. 


mmm €>t>e 

Courtesy General Electric Co. 

Small General Electric Electrostatic Electron Microscope. Circle Insert: Portion 
of fringe on mosquito wing taken with a General Electric instrument. 



Sis ^?l§/&,. \. ;V : ?ll||^||il||liiii 



Hi : * 



y . ^-. ...... ... ... ......:», v , .^.j, ...... ^ ■■.]:.^.::^i.^ 

Courtesy Dr. Stuart Mi 

Clostridium sporogenes, showing a mature spore within the remnants of the vegetative 

cell (X 40,000). 

Chapter 17 

Emission and Transmission Types of Electron Microscope 

The preceding chapters on electrostatic and magnetic lenses 
and on the electrostatic electron microscope have revealed the 
great flexibility of these electron-optical counterparts of light 
optical elements. We have encountered the equivalent of the 
simple microscope, or reading glass, in the single electrostatic or 
magnetic lens and also described combinations of electrostatic and 
magnetic lenses which represent in principle the compound optical 

The simple reading lens may be used for two purposes: 

(a) By holding it in a beam of sunlight we can focus on a 
screen an intense spot of light which is really an image of the 
sun; the same effect can be found by focussing the light from a 
an electric lamp. The feature of this experiment is that we get 
an image of the light source in each case. 

(b) We may also obtain on a screen an image of a window 
frame — an image of an object illuminated from an external source. 
In this case we are not primarily interested in the way in which 
the object is illuminated. 

When we come to deal with small (microscopic) objects we 
use several lenses which make up the compound microscope and, 
in this case, very great attention must be paid to the illumination 
of the object. 

We find the same two divisions in the development of electron 
microscopes. They fall into two classes: 

(a) Those in which the image is a reproduction of the surface 
of the cathode — the source of the electrons. Consequently this 
class, whether using electrostatic or magnetic lenses (or some of 
each) is known as the emission type of electron microscope. 
Electron microscopes of this type were the first produced. 




U O 

uj & 

-J u 

UJ — 



U If) 

r o 



.to ^ ^ 


^ Vj $J 




Comparison of Compound Light and Magnetic Electron Microscopes. 
(See Fig. 115, opposite) 

Elements Light microscope Electron microscope 

1. Source of illu- Sunlight or electric light. Electron source^ cold 

cathode or hot filament. 

2. Control of illu- Sub-stage condenser, Li. Converging action of the 
mination. magnetic field of the first 


3. Specimen sup- Microscope slide of glass Film of collodion about 


4. First image 
forming system. 

5. Accessory 

6. Medium. 

about 1 mm. thick. 
The objective, L 2 . 

The eyepiece or photo- 
graphic projector, L 3 . 

Air and glass. 

7. Viewing image. The eye directly. 

10 millimicrons thick. 

Converging action of the 
magnetic field of the 
second coil forming the 
first image. 

Converging action of the 
magnetic field of the 
third coil forming a sec- 
ond enlarged image of a 
portion of the first im- 

A very high vacuum. 

The eye (image pro- 
jected on fluorescent 

8. Method of 

9. Recording of 

10. Smallest particle 
in detail. 

Movement of lenses as a Alteration of the inten- 
result of judging the sky of the magnetic 
sharpness of the image fields in the coils as a 
with the eye. result of judging the 

sharpness of the image 
on a fluorescent screen. 

Photographic plate. 



Photographic plate. 

Less than 




(b) Those in which an object is inserted along the optic axis 
between the cathode and the screen and in which the image is 
a reproduction of this object, brought about by the interaction 
of the illuminating cathode ray stream and the material of the 
object. This class, whether using electrostatic lenses or magnetic 
lenses (or some of each), is known as the transmission type of 
electron microscope. 

In Figs. 21 and 22 we have indicated the two ways of making 
use of a compound microscope, viz., either to project the final 
(real) image on a screen or photographic plate or for observation 
of the final (virtual) image by the eye. The electron micro- 
scope is used only in the first sense; we have to depend on pro- 
jecting the image on a fluorescent screen, for focussing and 
observation, and on the photographic film, for reproducing 
pictures of the image. 

The term "simple" and "compound," as applied in micro- 
scopy, could be used of either the emission or the transmission 
type of electron microscope, depending on whether we are mak- 
ing use of one lens or more than one. 

The use of two solenoids in Fig. 85 (p. 164) for the purpose 
of producing a sharp image of the cathode spot on the screen is 
an example of a magnetic compound microscope. Since in this 
case a small spot is required on the fluorescent screen, the overall 
magnification has to be kept as low as possible, but full advantage 
is taken of the image-forming properties of the systems. 

Likewise, in Fig. 114a (p. 204) the two electrostatic lenses in 
the G.E. electron microscope comprise a compound microscope. 

During the first ten years of development (1932-1942), most 
emission type microscopes were of the electrostatic class, but 
those using magnetic lenses were also developed. Scott and his 
co-workers (St. Louis) described a two-lens magnetic type in 
1937 and in additional papers in 1939. They used the microscope 
to determine the distribution of metallic residues in biological tis- 
sues which were heated strongly on the surface of the cathode. 
Such emission types are usually used to determine some structure 
on the surface of the cathode. 



The basic patents for both electrostatic and magnetic electron 
microscopes were written in 1931 by Reinhold Riidenberg, and 
were assigned to his employers, the Siemens-Schuckertwerke, in 
Germany. Riidenberg later came to the United States, where he 
filed his patents in 1936 and 1937. Although these were broad 
in their specifications, and though Riidenberg never constructed 
an instrument, he has entered a claim to be considered the 
inventor of the electron microscope. 

History of the Compound Magnetic Electron Microscope 

The first emission-type microscope was built in Germany by 
Knoll and Ruska and described in the technical literature in 1932. 
It was of the emission type and served for the investigation of 
cold cathode surfaces; it was also used, for example, to form 
images of specially shaped apertures and wire mesh. 

Fig. 116 shows a diagram of the first Ruska-Knoll microscope 
as described in 1932. It was built at the High Voltage Research 

Kathode K 

der Elektronen 

Blende B 1 (Anode) 

Sammelspule Si 

elektrische Linse r^-^w-^ 

^ ! lT~~Blende B 2 

Einbaustelle fur die 

Fig. 116. 

Diagram of the Knoll- 
Ruska magnetic electron 

Beobachtungsschirm 5ch 

Fluoreszenzschirm r~ 






; _ . V( ,4,:; 

i " ' "^ 









'$ - 



All the reproductions on this plate are of cathode surfaces emitting 
electrons. These cathodes are the sources of the electron beam which 
is focussed on the photographic plate by means of single electrostatic 
or magnetic electron lenses. 

A, B, D, E and F are from a paper by Knecht [Ann. d. Physik, 20, 171 
(1934) ] . A has magnification X 9, the rest, X 31. D, E, and F are optical, 
electrostatic electron and magnetic electron microscope pictures, re- 

C is an electrostatic electron microscope picture of an emitting cathode 
obtained by C. E. Hall at Toronto in 1936. 

G, H and I are from a paper by Briiche and Mahl [Z. Techn. Physik, 
16, 240 (1935)], showing electrostatic electron microscope pictures of 
emitting thoriated tungsten cathodes, giving the distribution of the 
emission points. 


Institute in Berlin and followed in general outline the technique 
applied to the construction of high-voltage cathode ray tubes. 
This was only natural, since a cathode ray tube is essentially an 
inverted microscope, as we pointed out before, i.e., a microscope 
aiming at the smallest possible magnification of the cathode spot 
or of the anode aperture in front of it. 

In these early instruments the electron stream was obtained 
from a cold cathode in a gas discharge tube at the top of the 
microscope, according to the principles of cold-cathode emission 
in the presence of gas or air as described on p. 164 for the opera- 
tion of the Braun tube. A voltage of 50-80 kv was applied to the 
anode, which had a small aperture. Since the presence of gas 
was necessary in the discharge space above the anode and un- 
desirable in the main body of the microscope below the anode, 
air was admitted through a valve into the discharge space to pro- 
duce a pressure of about 10 microns (0.01 mm Hg) and a power- 
ful vacuum pump continuously exhausted the main bodv of the 
tube to produce a pressure of 1 micron (0.001 mm Hg). The 
anode thus served the dual purpose of accelerating the electrons 
and separating the low-vacuum space from the high-vacuum 
space below. A condenser lens, Si, served to gather the diffuse 
electron streams into a narrow beam, which was further con- 
stricted bv a second aperture stop and directed to a third aperture 
which could serve as the object. The solenoid below this aper- 
ture formed the magnetic objective lens, which produced an 
image on the fluorescent screen. This image could then be photo- 
graphed from the outside. For the purpose of focussing and 
visual observation of the image quality a side port was used in 
conjunction with a mirror and a fluorescent screen above the 
photographic glass plate. When the anode voltage and coil cur- 
rent were properly adjusted, the observation screen could be 
moved aside and a photograph taken. The beam currents ranged 
from 10 to 100 ma. 

A little later, during the same year, Knoll, Houtermans and 
Schulze built an instrument adapted to the investigation of oxide 
cathode surfaces which was operated at about 1000 volts. The 
axis of this microscope was horizontal and the components were 


arranged on an optical bench. Iron-clad coils were used instead 
of unshielded solenoids; but in this, and the first instrument de- 
scribed above, the coils were arranged outside of the main tube 
and in cardanic suspensions for ease of adjustment. 

The desire to obtain larger magnifications soon made neces- 
sary the construction of magnetic lenses with shorter focal lengths 
and the use of two lens stages in addition to the condenser lens. 
This brought about the construction of modern compound mag- 
netic electron microscopes, one by Ruska and one by Marton, 
in 1934. They were both of the transmission type and provided 
an object stage for the insertion of specimens. Marton was the 
first to investigate biological specimens with this instrument. He 
built a more elaborate type during the same year; this com- 
prised such features as air locks for changing the specimen with- 
out losing the vacuum in the main volume of the microscope, a 
mechanical stage of the cross-movement type, and an electrostatic 
shutter system for the control of exposure times. 

It may well be said that the two-stage magnetic electron 
microscope of Ruska, described in 1934, served as the prototype 
of all similar instruments which have been built since. Improve- 
ments were made continually by workers in many countries, and 
we may well marvel at the degree of perfection reached in modern 
designs when we consider that this art is only fourteen years old. 

The Ruska compound microscope of 1934, of which a photo- 
graph is shown in the plate on page 216, gave an overall magni- 
fication of X 10,000 and a resolving power of about 500 A.U. 
The objective and projector lens had a focal length of the order 
of 5 mm. The operating voltage ranged from 10 to 100 kv. 

In 1937 a compound microscope was described by L. C. 
Martin, R. V. Whelpton and D. H. Parnum in London, England; 
they incorporated an optical microscope in the magnetic micro- 
scope in order to be able to compare the relative image quality 
of optical and electron optical images. The minimum focal 
length of their magnetic lenses, which were of a special design, 
was 6 mm, while Ruska had obtained 3 mm. The results ob- 
tained were limited by technical difficulties later described by 



Knoll-Ruska electron micro- 
scope built in 1934. 

Courtesy L. 

View of 100-kv electron microscope at 
Stanford University. 

Early in 1938 a compound microscope was described by von 
Borries and Ruska which served as the prototype for the first 
commercial electron microscope produced by the Siemens Com- 
pany in Berlin. 

Also in 1937, the authors of this book initiated the construc- 
tion of the first compound electron microscope on the American 
continent by James Hillier and Albert Prebus, at the University 
of Toronto. This instrument was described in April 1939 and 
a picture of it is shown on page 80. The source of electrons was 
a tungsten filament bent into a hairpin shape. An operating volt- 
age of 45 kv was used and a resolving power of better than 200 
A.U. was obtained. Electronic magnification up to X 20,000 was 






Courtesy R.C.A. Laboratories 

The RCA Universal model electron microscope. Left to right, 
Drs. V. K. Zworykin, Perry Smith and James Hillier. 

Courtesy R.C.A. Laboratories 

The small magnetic electron microscope. 


reached and photographs taken with an exposure time of the order 
of five seconds. 

In 1940, von Ardenne described a new magnetic electron 
microscope in Germany and Marton and his co-workers, M. C. 
Banca and J. F. Bender, developed the first R.C.A. electron micro- 
scope under the direction of V. K. Zworykin; this developed into 
the first commercial R.C.A. microscope, following the design of 
Hillier and Prebus (Toronto). Prebus, who joined the faculty 
of Ohio State University, published a description of his new 
instrument in 1942 . Zworykin and Hillier built a small-scale 
magnetic microscope in 1943. 

In addition to the commercial instruments installed in various 
laboratories and those already mentioned, electron microscopes 
of this type have been constructed by C. E. Hall at the Research 
Laboratories of the Eastman Kodak Co. at Rochester and by Pro- 
fessor Marton at Leland Stanford University. A new and im- 
proved microscope has recently been built at the University of 
Toronto; it was first used in June, 1944. 



Courtesy (joodyear Tire ana Kuuoei 00. 
Rubber research with the electron microscope. (Pencil drawing by T. Kautzky.) 

Courtesy University of Toronto 

View of the electron microscope now in use at the University of Toronto. This instrument, 
built in the Physics Department workshop, was completed in 1944. 

Chapter 18 


A detailed description of the design, construction and opera- 
tion of this latest Toronto instrument will now be given, for no 
other reason than that this is the instrument with which the 
authors are most familiar. With this treatment will be inter- 
woven comparative remarks referring to other modern instru- 
ments in order to give a general description of the fundamental 
technique as it is known and practiced today. 

This instrument was designed by Drs. Newman and Watson 
in the light of experience gained with the microscopes built at 
Toronto in 1938. The makers and users of the R.C.A. model B 
microscope were also consulted during the design of the Toronto 
model. The microscope proper was constructed entirely in the 
workshop of the McLennan Laboratory, University of Toronto, 
and was first put into operation in the summer of 1944. 

Description of the Instrument 

Fig. 117 shows the detail. G is the electron gun. A tungsten 
filament, heated by a D.C. current of about 3 amps is maintained 
at —45,000 volts with respect to ground. S is a cathode shield 
at the same potential as the filament; it produces an electrostatic 
lens which f ocusses the electrons leaving the filament. The region 
of minimum cross-section of the beam, the cross-over, see Fig. 
103 (p. 193), may be considered the source of illumination for 
the microscope. A is the anode, which is connected directly to 
the body of the microscope and is at ground potential. 

Electrons passing through the central hole in the anode are 
concentrated on or just above the object or specimen by the con- 




denser lens C. A diaphragm of about 0.5 mm diameter, (not 
shown), within the condenser lens serves to limit the maximum 
aperture or semi-angle of the cone of electrons falling on the part 
of the specimen under observation. 

Some of the electrons which leave the object are focussed by 
the objective lens O on the intermediate fluorescent screen I.S. 

Fig. 117. 

Diagram of the 1944 Toronto 
magnetic electron microscope. 

A hole in this intermediate screen permits some of the electrons 
to pass through and be focussed by the projector lens P on the 
bottom fluorescent screen B.S. The screen material is zinc-cad- 
mium sulfide. Here there is formed an enlarged image of the part 
of the first image corresponding to the hole in the plate I.S. 
When this lower screen is turned up, the image is formed on the 



photographic plate P.P. The camera allows a 2 x 10-inch photo- 
graphic plate to be racked through step by step and the electron 
beam can be masked to expose an area of, say, 1% x 2 inches, and 
thus gives six exposures on each plate. Provision is also made 
for the use of roll film. 

The Magnetic Lenses 

The lenses of the latest Toronto electron microscope consist 
essentially of the core, coil and pole pieces. 

The iron core of the lens ends at conically ground faces F, 
Fig. 118. The pole-pieces of the lens, Ps and Pn, are joined by 


Fig. 118. Detail of a magnetic lens and two types of pole pieces 
in the 1944 Toronto electron microscope. 

a narrow neck of iron N, so that they can be machined en bloc, 
thereby assuring a high degree of symmetry. The pole-pieces 
are made of Armco iron. This type of pole-piece at the top was 
designed by Prebus. The outer faces, F x , are accurately ground 
to fit the faces, F, of the lens core. This design was produced in 
order that developmental work could be carried out on the pole- 
pieces. New pole-pieces can be machined and fitted to the lens 
with the minimum of structural alterations. 



The mode of image formation in the electron microscope is 
rather different from that of the light microscope. Electrons 
from a heated filament are focused on the object by a magnetic 
condenser lens. Some electrons are absorbed or collected by the 
object and others are scattered in varying angles both with and 
without loss of speed. Only those electrons passing through or 
close by the object without loss of speed are brought to a focus on 
the photographic plate (See Fig. 119). 

*- Psthofe/ecfrons 

scattered us/thou t /oss of speed 

^* Path of e/ec frons 

r\ scattered urith loss of speed 


Fig. 119. Diagrammatic view of the path of electrons through a magnetic lens. 

If then, there is a particle of matter at O, it will scatter elec- 
trons out of the beam and leave a deficiency of electrons in the 
image plane at I. This deficiency can be enhanced, and thus 
also the contrast in the image, by using a diaphragm, D, to remove 
electrons scattered through angles greater than a certain value. 
Thus in the positive prints, dark areas correspond to the presence 
of matter in the object plane. 

In the Toronto microscope diaphragms can be placed at the 
center, or below, any or all of the lenses and aligned from outside 
the instrument while it is in operation. Good results have been 
obtained with an objective diaphragm placed immediately below 
the lower pole-piece. In this position the diaphragm sheet can be 
pierced by one hole of 2 mm diameter and three holes of 0.05 mm 
diameter. The alignment of the lenses is carried out with the 
large hole approximately centered, and then one of the smaller 


apertures can be moved into place. If this one becomes dirty, 
another aperture can be moved into position without breaking the 

Since the varying magnetic fields produced by transformers 
and motors in the laboratory would deflect the electron beam 
during photographic exposures, and thus blur the image, the 
beam is shielded from such fields by mu-metal shields. 

The optical performance of the microscope will depend on 
the stability of the electrical supplies, both for the gun and for 
the various lens coils. 

Structural Features 

The body of the microscope must be designed so that the 
microscope can be aligned and then clamped to keep the lenses 
rigidly in position. The object holder must be capable of con- 
trolled motion in two directions, mutually perpendicular, so that 
different parts of the specimen can be examined. The whole 
microscope must be protected against vibration, for even a minute 
vibration of the specimen relative to the objective lens aperture 
would blur the final image. These conditions must be met while 
maintaining a high degree of vacuum within the microscope. 
It is also desirable that the interior of the microscope be readily 
accessible for cleaning and changing the diaphragms from time 
to time. 

The filament can be raised or lowered relatively to the cathode 
shield and moved laterally in two directions at right angles. The 
cathode shield can be moved laterally with respect to the anode 
aperture. The condenser lens and all parts above it can be shifted 
laterally with respect to the objective; the objective and all parts 
above it can be moved and also tilted with respect to the projector 
lens. These movements can all be made from outside the instru- 
ment while it is in operation. 

The object stage consists of two flat bronze carriages, the 
upper moving on the lower and the lower moving on a plate fixed 
to the objective lens. The two carriages move at right angles to 
each other on ball bearings in F-grooves lubricated slightly with 
Celvascene light grease; they are driven by 0.5-mm pitch mi- 


crometer screws. The carriages are held down and against the 
micrometers by rubber bands. This results in a smoother motion 
than is readily obtained with metal springs. A conical taper is 
cut in the upper carriage, and into this the object cartridge is 
lowered by two nylon threads. The use of the conical taper 
seats the object cartridge very accurately for depth and holds it 
rigidly. When a plunger, PL (Fig. 117) on the side of the in- 
strument is withdrawn, the object cartridge is first raised ver- 
tically and then turned into a horizontal position so that the 
object air lock can be closed by screwing in the seal, Si. 

Vacuum Arrangements 

The pressure within the microscope must be maintained at a 
value below 10" 5 mm of mercury in order that the mean free path 
of the electrons shall be longer than the optical path they have to 
travel, and also in order that gas discharges shall not occur within 
the electron gun. It is further desirable that this low pressure be 
attained as rapidly as possible when the microscope is put into 
operation, or after the photographic plate or the specimen is 

It is important that materials facing the interior of the micro- 
scope shall have a low vapor pressure. In the Toronto micro- 
scope the seals between sections are made by neoprene gaskets. 
Rotating members such as those for moving the diaphragms and 
the object-stage micrometers pass through featheredge neoprene 
seals of the type developed for use in the cyclotron. Where 
tilting or traversing motions have been introduced the vacuum is 
maintained by the use of Sylphon bellows. The glass insulators 
in the electron gun are sealed in with apiezon W wax, and the 
only use of grease is for lubrication of some of the moving parts; 
this grease is Celvascene light vacuum grease. 

For rough checks of the pressure a discharge tube is sufficient. 
The absence of a visible discharge indicates that the working 
vacuum has been obtained. For more precise measurements a 
good McLeod gauge or ionization gauge is used. 

Both the object and the photographic plate can be introduced 
through air locks which are evacuated by a Hyvac pump, so that 


the specimen or photographic plate can be introduced and the 
microscope brought to the operating vacuum in less than two 
minutes. This can be done without switching off the high volt- 
age on the gun. The main pump is a Distillation Products MC 
275 oil diffusion pump, backed by a Hypervac pump. Three 
2-inch diameter tubes connect the microscope, through sylphon 
bellows, to the oil-diffusion pump. 

Associated Equipment 

The associated equipment consists of that for the high 
voltage, and the circuits for providing the lens currents. Since 
the focal length of a magnetic lens is directly proportional to the 
square of the electron velocity, and inversely proportional to the 
square of the magnetic field strength, it is very important to main- 
tain the accelerating voltage applied to the electron gun and the 
currents through the lenses as nearly constant as possible, because 
the speed of the electrons entering the object depends on the 
accelerating voltage, and the field strengths within the lens depend 
on the currents through their windings. 

The stabilities of the power supplied to the R.C.A. model B 
microscope are listed below: 



Stability (%) 

Acceleration potential 

55 kv 


Condenser lens 

20-120 ma 


Objective lens 

20-130 ma 


Projector lens 

25-125 ma 


The methods for obtaining such a high stability in voltage and 
currents have been discussed by Vance. At Toronto, methods 
simpler in conception, if more wasteful of space, are used. The 
lenses have between 2,000 and 3,000 turns of copper wire and 
resistances of the order 20 ohms. The currents are supplied by a 
24-volt storage battery, and are controlled by two rheostats in 
series with each lens, one for coarse adjustments, and the other 
for fine adjustments. The general method is shown in the circuit 
diagram of Fig. 120. 

The transformer, T x , steps the voltage up to 45 or 60 kilovolts 
(peak), the kenetron, K, rectifies the voltage, and the resistance 



X R 3 R 


Y& rT 


Fig. 120. Circuit diagram of power-supply for the 1944 Toronto electron microscope. 

IS a 

capacity filter R 4 , d, C 2 reduces the ripple to a few volts. R 
protective resistor to prevent Ci from discharging destructively 
through the microscope if a gas discharge occurs. A fraction 
Ri/(Ri + R 2 ) of the high voltage is amplified by the D.C. am- 
plifier (D.C.A.) to drive the tube T 2 , so that the fluctuations in 
its anode potential tend to counteract fluctuations in the potential 
difference between X and Y, thus preserving a more constant 
voltage across the microscope, M. Arrangements of this sort have 
been discussed by Hunt and Hickman, and by Parratt and Triska. 






* « 

# ? 



* *f 

Courtesy Dr. T. F. And 

E. coli and virus (X 21,000). 


Bacteriophage and coliphage (X 53,000) 

Chapter 19 

Resolving Power and Magnification 

While the phenomenon of diffraction and consequently the 
wave length of the light used determines the limit of the resolving 
power of the ordinary optical microscope, in the case of the 
electron microscope, we cannot reach anything like the theoretical 
resolving power set by the wave length assigned to the electron 
beam. We are limited rather by our inability to get rid of aber- 
rations and structural deficiencies. In other words, our limits are 
not set by the equation: d = ilo (p. 73), as, in this case, the 
value of l is very much smaller than the dimension which we can 
attain experimentally as the limit of resolution of the electron 
microscope. The reason for this is that the aberrations, par- 
ticularly the spherical aberration produced by magnetic lenses, are 
large unless the angular aperture of the illuminating beam is kept 
small, e.g., of the order of 10" 3 radian, as compared to 1 radian in 
light microscopes. 

Theoretically, for accelerating voltages of the order 50 kv 
and apertures of the order 10~ 3 to 10" 2 radian, d will be approxi- 
mately 10 to 20 A.U. Such resolutions have been claimed by 
some in practice. 

The Practical Limit of Useful Magnification 

If, to err on the safe side, we assume that the smallest particles 
or details resolved by present-day electron microscopes are about 
30 A.U. in linear dimension (i.e., three ten-millionths of a centi- 
meter), we can estimate the magnification required of the micro- 
scope. As a matter of practice, all measurements must be made 
by observations of an image impressed on a photographic emul- 
sion (plate or film) . We thus have to consider the limitations of 




Courtesy University of Toronto 

Stereoscopic views of particles caught on single fiber. 

Courtesy University of Toronto 

Sodium chloride crystals caught on a rubber fiber. 




2. ,3 


Courtesy University of I oronto 

Illustrating how particles collect more abundantly on smaller fibers. The diameter 
of the fibers are recorded at the top of diameters 1.0fi and 3.3^. 

the photographic film — the resolving power of the photographic 
emulsion. An ordinary value for this is about 70 lines per milli- 
meter. This means that such a photographic film will not sepa- 
rate lines which are closer together than l/70th millimeter, or 
l/700th centimeter, or record images having linear dimensions 
smaller than this. Consequently, it is quite apparent that a 
particle which has an overall linear dimension of 30 X 10 -8 
cm must be magnified to a dimension of at least l/700th cm in 
order to be distinctly outlined on the film; that is, the particle 
must be magnified by an amount given by 1/700 divided by 
30 X 10" 8 , or about 4760 times, to give an image on the film. 

Now, for the photograph to be useful, we must be able to see 
the finest detail in an image with the naked eye. The naked eye 
cannot see particles as small as 1/700 cm; the limit for the eye is 



about 1/100 cm. Therefore, the above photographic image must 
be magnified at least seven times in order that the eye may see it. 
Thus the total magnification will have to be approximately 
4760 X 7, or 33,320. 

"As the microscope itself is capable of magnifications from 
1200 to 23,000 diameters, the useful magnification for photo- 
graphic purposes falls well within its range. Considerable latitude 
in procedure is available for obtaining high total magnification for 
naked eye viewing. Pictures may be made with the microscope 
at 23,000 diameters and slightly enlarged optically to a convenient 
size for study, or they may be made as low as 5000 diameters and 
enlarged ten or fifteen times. Experience has shown that the 

Courtesy University of Toronto 

Diatom (Synedra delicatissima) X 1,600. Collection of diatoms to show distortion of 
the image of one diatom taken in different parts of the field. 

latter procedure is not too satisfactory; because, in order to get 
a photograph capable of such enlargement, extreme care must be 
taken in the development and fixing of the plate to avoid excessive 
clumping of the silver grains and hence loss of resolution. The 
former procedure is also subject to objections, chiefly because the 
intensity in the final image is so low that accurate focussing is 
difficult and the exposure time becomes so long that the electrical 
stability of the instrument cannot be relied upon to give the 
ultimate resolving power. 

"It has been found most satisfactory to photograph at a micro- 
scope magnification of about 9000 diameters for general work, 
and if extremely high resolution is needed because of the detailed 
structure of the specimen (particularly for thin metallic films) 
a series of photographs made at about 17,000 diameters will give 


excellent results. Sharp enlargements with over-all magnifica- 
tion ranging from 50,000 to 150,000 diameters can regularly be 
made from these plates without using special fine grain develop- 
ment technique." (Picard and Duffendack, /. App. Physics, 14, 

291 [1943]). 

Since it is sometimes desirable to compare electron micro- 
graphs with light micrographs at nearly the same magnification, 
this and the preceding considerations make it desirable that the 
microscope shall be capable of producing magnifications in the 
range X1500 to X 20,000. This can be achieved by using an 
objective lens which will give a magnification of X100 with a 
projector lens giving a maximum magnification of X200, or by 
using both lenses with a magnification of about X 140 each. 

The magnification produced by a magnetic lens is given 
very approximately by v/u, where u is the distance from the 
object to the point of maximum axial magnetic field and v is the 
distance from the image to the maximum axial magnetic field. 
It is desirable to keep v as small as possible so that the microscope 
shall not be too large and unwieldy. At the same time if a very 
small value is chosen for u it will bring the object within the 
pole-pieces of the objective lens where it will not be possible to 
move it laterally over any considerable distance. The field avail- 
able for investigation will be reduced; in any case the magnifica- 
tion will not have increased as rapidly as v/u. Another reason for 
preferring larger values of u is that the focal length of the lens 
may then be made larger, and the pole-pieces in the lens may be 
made larger with consequent increase in the degree of accuracy 
in the machining. 

For these reasons the values of v for the objective and projec- 
tor stages of the Toronto microscope have been made about 40 
cm. An advantage of these dimensions is that the necessary 
optical adjustments can be made by the operator while the 
fluorescent screens are in full view. This facility of adjustment 
has also been maintained in the choice of the distances between 
the electron gun and the condenser lens, and between the con- 
denser lens and the object stage. In a well made magnetic lens 
there is a straight path along which electrons can travel without 


being deflected; this is called the axis of the lens. Images of points 
off the axis of the lens rotate about the axis as the current through 
the lens coil is changed. In order to produce good electron 
micrographs it is necessary to bring the axes of the three lenses 
into coincidence and to bring the source of electrons onto this 
axis. This is called aligning the microscope. 

Since it is often desirable to determine the size of small objects 
photographed in the electron microscope it is necessary to know 
the magnification produced by the instrument. Two general 
methods are available. If we can measure the displacement of the 
object stage (and hence the object) necessary to move the final 
image through 3, 4 or 5 cm, the overall magnification can be cal- 
culated from 

displacement of image 
displacement of object 

In the new Toronto microscope, circles of radii 1, 2, 3, 4, and 
5 cm are drawn on the bottom fluorescent screen, and the microm- 
eters that move the object stage are calibrated to read to 0.0005 
cm; displacements can be estimated to one-tenth of this. At mag- 
nifications above X 5000 this method becomes insensitive. 

The second method is to introduce into the microscope 
objects of known size and then to measure the size of the image on 
the developed photographic plate. For this purpose C. J. Burton, 
R. Bowling Barnes, and Rochow have used the invar replicas of 
diffraction gratings, and Fullam has used microscopic glass spheres 
of predetermined size. 

Above: Ordinary electron microscope picture of a calcium stearate grease (X 20,000), 

Below: Shadow electron micrograph of sample of a similar grease (X 35,000); 

shadows are cast by chromium molecules. See also page 86. 

*<■ jr 

- m >«sit f 

Vibrio schuylkiUiensis (X 22,500). 

Courtesy Dr. Stuart Mudd 

Chapter 20 


The general problem of electron micrography falls into three 
main parts (1) the preparation of the specimen, (2) the produc- 
tion of the micrograph, (3) the interpretation of the micrograph. 
These will be considered separately. 

Specimen Preparation 

The methods of specimen preparation employed are very 
diverse and will have considerable influence on the quality of the 
final micrograph. There is also the possibility of introducing 
artifacts, or of modifying the specimen during the mounting and 
introduction into the microscope. The methods at present 
employed are: (1) direct mounting; (2) mounting on, in, or 
between formvar or collodion films; (3) replica techniques; (4) 
mounting in water. These methods will be illustrated by con- 
sidering special cases. 

Direct Mounting. Certain biological specimens, such for in- 
stance as the wing of a midge (chironomidae) can be mounted by 
placing them on a 200-mesh screen for support, or by cementing 
them to the end of the object holder. Again, in studying the 
action of fibers in collecting air-borne particles, the fiber may be 
mounted across the end of the object cartridge, the particles 
blown past it and the cartridge then put into the microscope for 
examination. The advantage of this method is that it avoids the 
scattering produced by a film and gives rather better contrast in 
the image than is usually possible with specimens mounted on a 



The Use of Films. There are many variations of this method. 
Three are considered here. A method often used in mounting 
bacteria is to produce a formvar film and transfer it to the top of 
a 200-mesh screen. The organisms to be investigated are sus- 
pended in water and a drop of the suspension is placed on top of 
the film and allowed to evaporate. If the suspension is concen- 
trated, the residual liquid may be removed with a filter paper after 
a little of the specimen has settled out. If the suspension is dilute 
several drops may be allowed to evaporate in succession. This 
method is commonly used in mounting bacteria and may be used 
in mounting colloids dispersed in water. 

In another method, a glass microscope slide is dipped into a 
0.1 per cent solution of formvar in ethylene dichloride. On re- 
moval, the ethylene dichloride evaporates, leaving a thin film of 
formvar on both sides of the glass. If the film is cut with a 
needle it may be floated off on water and then mounted on a 200- 
mesh screen and used as before; or the specimen may be mounted 
or dispersed on the film while it is still attached to the glass; it is 
then cut into suitable sizes, floated off on to water and mounted 
as before. In this method a larger area of film is available for 
mounting and it is well supported on the glass. Finally, the sub- 
stance can in some cases be dispersed in the solution from which 
the films are to be made, and it then appears encased in the film 
and is mounted as before. 

Specimens mounted on a film may be covered with another 
film. This is not usually necessary and in any case it leads to an 
increase in electron scattering in the microscope and a decrease in 

Films of collodion formed on water can be used in place of 
formvar films. Both are used at Toronto, but our preference is 
the latter. 

Replica Techniques. The surfaces of solids cannot be di- 
rectly examined in the transmission type microscope since even 
the thinnest films of most solids are in general opaque to the elec- 
tron beam. However, a formvar copy or replica of the surface 
can be made and examined in the microscope. In fact, if stereo- 
scopic micrographs are made, the depths of scratches and cavities 


in the surface can be determined. In a variation of this tech- 
nique, a polystyrene replica (a negative) of the surface is made 
and from this a silica replica (a positive) is produced; the latter is 
examined in the microscope. While replicas were used originally 
for the study of etched metal surfaces the technique is now being 
extended to the study of many other types of specimen. 

Mounting in Water. A cell for mounting substances in 
water and examining them at atmospheric pressure has been de- 
scribed. At the time of writing, micrographs by this method 
have not been published, but owing to the speed and magnitude 
of the Brownian movement, micrographs can be obtained of small 
particles only if they are attached to the containing films, and 
even then good contrast can only be expected with particles of 
high specific gravity, such as metallic colloids. 

After the specimen has been mounted by any of the above 
methods it is good general practice to make a light-microscopic 
examination of the dry specimen in order to see if the deposit on 
the film is of suitable density and free from dust and foreign 

The Production of the Micrograph 

Before the specimen is examined in the electron microscope 
it is desirable to align the microscope, if this has not already been 
done. The minimum intensity of illumination consistent with 
ease of focussing is then chosen by adjusting the filament current 
and the specimen introduced. The following methods of micro- 
graphy are then available: (1) direct illumination; (2) dark-field 
illumination; (3) stereoscopic micrography; (4) motion-picture 

Direct Illumination. The major factors to be considered 
here are the choice and measurement of the magnification, the 
choice of intensity and time of exposure, the use of diaphragms 
and the observation of features not readily reproduced in a photo- 

Generally speaking, the minimum magnification is used in the 
examination of small particles when size distribution and structure 



I TqZ. I 

i >.'etr-o:. iT nainei (No 





,-■,:■:; ■•:t.':'i 


A '^7 l .4t§fp 

- fr' 



\; £v. 


"-/' . .*.:. 


- /. '". 

.// ■'■ 





7 a*sR> 

, ■ - 0: 

Vy / 


v ... j^-»v-,.v*r 

Courtesy Dow Chemical Co. 

Electron micrographs of etched surface and interior structures of a human tooth. 
Polystyrene-silica replicas (X 1,200). 



Courtesy Dow Chemical Co. 

Polystyrene-silica replica of external surface of human tooth enamel (X 5,300). 

No etch. 

Courtesy Dow Chemical Co. 

Polystyrene-silica replica of the interior structure of rabbit bone, polished and 

etched (X 5,300). 


are to be found, for the smaller the magnification the greater is 
the area of the object observed and hence the greater the informa- 
tion provided. However, the desired resolution in the micro- 
graph and the grain size of the photographic plate set a lower 
limit to the permissable magnification. 

The image is usually focussed at maximum intensity as ob- 
tained by changing the condenser current until the electrons from 
the gun are brought to a focus in the plane of the object. The 
angular aperture of the illuminating beam is then given approxi- 
mately by 

radius of diaphragm in condenser lens 
distance from condenser diaphragm to object 

The angular aperture of the illuminating beam is then reduced 
by increasing the condenser current slightly, and the exposure is 
made. For particles mounted on a film, particularly the less dense 
particles and the thicker biological specimens, the use of a dia- 
phragm immediately behind the objective lens improves the 

. The visual observation of the image on the bottom fluorescent 
screen will sometimes provide information not readily recorded 
on a photographic plate, though the motion-picture method 
might show it. For instance, the specimen may change its form 
under electron bombardment; it may evaporate, signs of dehydra- 
tion or drainage may appear, and a breakage of the film may lead 
to mechanical forces being exerted on it, the results of which can 
be seen on the screen. Sometimes parts of the specimen may 
move or vibrate due to their becoming charged by the electron 
beam, with the consequent result that they are repelled t>y the 
following electrons. The movement and vibration may give 
further information as to the shape, rigidity, and structure of the 
specimen. Crystalline material in the specimen will sometimes 
give rise to spots and bands of light on the fluorescent screen 
which move with small changes in the objective current. This 
may be used as a test for the presence of crystalline material in 
some cases. 

Dark-field Illumination. This method, analogous to the 
method of the light ultramicroscope, has been used by von Ar- 


denne and by Levy, the latter reproducing some interesting 
pictures. At present the resolving power of the electron micro- 
scope used with dark-field illumination does not show any great 
improvement over that with direct illumination. It is probable 
that there is room for development here. 

Stereoscopic Micrography. The relatively large spherical 
and chromatic aberrations of the magnetic lens have made it 
necessary to work at angular apertures of the order 10" 3 and 10" 4 
radian. This is very much less than the value used in light micro- 
graphy, and consequently the depth of focus in the electron 
microscope is relatively large. In practice, particles separated by 
lOfju along the axis of the microscope can be considered in focus. 
If one micrograph is made and the object then tilted through an 
angle of say 10° and another micrograph is made, and the two are 
viewed by the right and left eye respectively, an appearance of 
depth is given which is qualitatively useful. Further, the parallax 
between the micrographs can be measured by the aid of the instru- 
ments used in contour plotting from aerial photographs and 
quantitative measurements made in the third dimension. 

With crystalline material, such as fine metallic particles 
deposited on a collodion film, this method may break down, since 
the appearance of the field may change considerably when the 
specimen is tilted. The reason for this is the interference between 
the reflected electrons that produces the light and dark spots and 
bands previously mentioned. Approximate measurements of the 
heights of particles can still be obtained, however, if a metal such 
as chromium or uranium is evaporated on to the supporting film, 
obliquely, so that the object under study casts a shadow — a region 
in which no chromium is deposited. From the length of this 
shadow, as observed in the electron microscope, a minimum 
height for the object can be deduced. 

Surface irregularities of the order of 30 A.U. have in this way 
been detected on collodion films. 

Motion-picture Micrography. This has been reported by von 
Ardenne in attempting to secure pictures of the progress of 
chemical reactions. 



Interpretation of the Micrograph 

An electron micrograph taken by direct illumination gives 
information only about the shape, size, and structure of the speci- 
men in the two dimensions of the object plane. 

Owing to the great depth of the focus of the electron micro- 
scope the objects of Figs. 121 (a) and (c) would both produce 
images of the form of Fig. 121 (b). The two objects could, 
however, be distinguished by the aid of stereoscopic examination. 
This effect should be kept in mind when examining the electron 
micrograph of clumped particles, since the particles may build up 
off the plane of the film in "chains" and "trees." 




o b c 

Fig. 121. Interpretation of micrograph. Objects (a) and (c) 

would each produce an image (b). 

This is perhaps the most difficult point to grasp in the inter- 
pretation of electron micrographs and is more difficult for 
observers used to light micrographs in which a thin slice of the 
object is in focus and the adjacent planes are blurred. 

In the photographic plates (negatives) the blackening is a 
function of the total mass that the electrons have had to traverse, 
(neglecting the effect of scattered electrons) and cannot usually 
be interpreted in terms of thickness unless uniform density of the 
material can be assumed. If the specimen is known to possess 
uniform thickness, blackening indicates regions of lower density. 



An important step in the interpretation of any micrograph is 
the detection of artifacts, that is, appearances in the micrograph 
which do not correspond to actual structures in the naturally 
occurring substance. They may be introduced by the method of 
specimen preparation, or by the method of micrography. The 
general method of detecting artifacts is to change the factors 
which may produce them and look for changes in the micrograph. 
A few of the more common are listed below. 

Narrow light bands around particles. These are due to chro- 
matic aberration and diffraction, and might be misinterpreted as 
sheaths in pictures of bacteria if care is not taken. 

Bands or spots. These appear in images of crystals, and are 
most readily apparent if no diaphragm is used in the objective; 
they shift with a change in focus. 

Fig. 122. Differential blackening of images from objects of 
uniform density due to twisting or packing. 

Blurring of edge of image. This may be due to improper 
focus, lack of resolution, change of focus during exposure or a 
shift of the object during exposure. Another cause sometimes 
observed, e.g., with flowers of sulfur, is evaporation during ex- 
posure. Shrinkage, dehydration or spontaneous changes of shape 
occur with some materials. Shrinkage and dehydration some- 
times occur with biological specimens, and a spontaneous change 
of shape from a jagged to a smooth shape has been observed with 
tgg albumen under intense illumination. These effects are best 
detected by visual examination of the image on the fluorescent 



Distortion. This can occur in some microscopes, but it ap- 
pears to be negligible in the new Toronto microscope at magnifi- 
cations above X2500. It is detected by measuring the size of 
particle photographed first at the center and then at the periphery 
of the field. Visual observation of the straight edge of a sodium 
chloride crystal, as the image is moved across the bottom 
fluorescent screen, provides another method of detecting its 

Deformation caused by method of mounting. Salt crystal- 
lizing from thin layers of salt solution on f ormvar produces plate- 
lets rather than cubic crystals. Mechanical deformation of the 
specimens is not very likely, as most specimens are handled sus- 
pended in water. The possibility that small particles may take up 
preferred orientations on the formvar film should not be over- 

It has already been pointed out that photographic enlarge- 
ment may be necessary before all the detail in the micrograph is 
readily visible. At Toronto, the 2 x 10-inch negatives are usually 
examined in a Leitz projector giving a further magnification of 
Xl5, in order to decide which fields shall be enlarged. This 
procedure also facilitates rapid comparison of micrographs and 
provides an image in which particle-size distribution counts can 
be made without the production of a large number of standard- 
ized photographic enlargements. 

In conclusion, we may say that an experienced observer can 
make accurate statements as to size, shape, and structure (which 
will include size distribution) down to magnitudes ten times 
greater than the resolution limit of the microscope, and that the 
artifacts do not impose serious limitations on the use of the elec- 
tron microscope. 

This type of information, taken in conjunction with results 
from bacteriological investigations or in chemistry from colloidal, 
x-ray, electron diffraction or electron analyzer experiments, will 
go far to systematize and develop the science of small particles 
and the industries which are based on them. 

That the electron microscope is limited to the measurement of 
size, shape, and structure and also that the results it produces must 


be considered in their relation to other experiments on the speci- 
men should be emphasized. The electron microscope is not a 
cure-all for research difficulties, but in its own field it is unrivaled 
and it will often lead to the clarification of difficulties in other 
fields of research. 

Trends in Electron Microscopy 

In the design of modern compound electron microscopes of 
the magnetic type we may notice two trends. One is the produc- 
tion of a "Universal" microscope which can be used at magnifica- 
tions between 2000 and 20,000 diameters for ordinary, dark-field 
and stereoscopic work. Such microscopes often contain the neces- 
sary equipment for electron-diffraction investigations. The other 
tendency is to the production of small, simplified microscopes 
restricted to one or two values of magnification and without the 
electron diffraction attachments. The larger model can be used 
for any research in the field of electron microscopy; the smaller 
is intended for production checking, diagnosis or other more 
restricted uses. 

At the same time the universal model has been modified to 
perform new functions; stereoscopic photography, motion-pic- 
ture photography and electron-diffraction techniques are 
examples of this. 

Research is being made upon the lens aberrations and already 
basically new procedures are being devised to reduce or nullify 

The basic technique has also been turned to the production of 
a different type of instrument, the electron analyzer, which can 
be used for the detection of chemical elements in a previously 
observed small region of a sample. Here we have a further 
development in the experimental methods of studying small par- 


Courtesy Dr. Stuart Mudt 

Staphylococcus flavocyaneus: The first indication of , intracellular granules (X 21,000). 

Chapter 21 


All the important practical applications of the electron micro- 
scope are due to two great advantages which it has over the optical 

(1) increased resolving power, enabling one to see and photo- 
graph fine particles and fine detail; 

(2) greatly increased depth of focus enabling one to get use- 
ful magnification in three dimensions. 

The disadvantages of the instrument are chiefly due to the fact 
that all specimens have ultimately to be exposed in a vacuum and 
consequently will have to undergo changes brought about by 
drying and evacuation. In addition, the micrograph cannot be 
taken without exposing the specimen to rather an intense electron 
beam for the considerable length of time required for focussing. 

The electron bombardment may induce some changes; and 
under certain circumstances of mounting, the specimen may 
become electrically charged. As a consequence it is not advis- 
able to rush into a rash interpretation of any particular micro- 
graph. However, there is no doubt that the electron microscope 
has made a place for itself and gives promise of striking develop- 
ments in the future. 

Up to the present (1945) the magnetic electron microscope 
has been used extensively in many fields. 

Fine Details of Natural Objects 

. The determination of fine details in the structure of various 
natural objects, such as diatoms, pollen dust, the trachea and the 
wings of insects was one of the earliest applications of the electron 
microscope. As diatom shells are common test objects for optical 




Courtesy Columbian Carbon Co. 

Above: Thiokol latex particle as seen in the electron microscope (X 12,500), 
Below: The same as seen through light microscope (X375). 


microscopes, it was natural that they should have been used for 
some of the first objects for these investigations. Illustrations of 
various kinds of diatoms are shown on pages 24, 38, 40, 76 and 
234. The Department of Physics of the University of To- 
ronto has had the kind cooperation of Dr. Paul S. Conger of 
Washington, D. C, who has supplied a great variety of diatoms 
for electron micrographs. 

As shown in the picture of the wing of a midge (Frontispiece) 
the electron microscope has revealed many new, unexpected 
structures in animals and plants. 

Sizes and Shapes of Small Particles 

Carbon Black and Fillers. Historically, the first commercial 
application of the electron microscope was the determination of 
the size and shape of small particles, such as the particles of carbon 
black used for reinforcement in rubber, or the particles of pig- 
ments used in various paint products. Pioneer work in this field 
has been done in the laboratories of the Columbian Carbon Com- 
pany of New York. Considerations raised by these carbon black 
pictures have revolutionized the theories of the role of rubber 
reinforcing agents. These particles take part in a physico-chemi- 
cal reaction with the rubber, and the intensity of the effect de- 
pends on the total area of contact between the carbon particle 
surface and the rubber. These carbon particles are apparently 
spherical and some of them are as small as 40 m/i (4 x 10" 6 cm) in 
diameter. If we should take a cube of sugar measuring one centi- 
meter on a side and grind it into small spheres of this diameter, 
the surface exposed by the sugar would increase from 6 sq cm to 
over 800 sq ft. Thus we can easily see how important considera- 
tion of surface areas may be. 

Clays. The study of various kinds of clay has been of very 
great interest to the soil chemist, i.e., to agriculture, and to the 
colloid chemist who has been in contact with the many commer- 
cial applications of clay, e.g., its use in ceramics, in china man- 
ufacture and for paper coating. A very great deal of atten- 
tion has been given to the colloidal properties of clay and naturally 
one of the main branches of this study has had to do with the 



Courtesy A. Prebus, Ohio State University 

Kaolinite crystals (X 29,000). 



Courtesy A. F 'rebus, Ohio State University 
Halloysite crystals (X 44,000). 


determination of the size and shape of the ultimate particles which 
go to make up the clay aggregate. Probably the main ingredients 
from this point of view are the clay minerals. 

Specific study of clay minerals dates back to about 1930, al- 
though it was shown by C. S. Ross in 1927 that crystalline clay 
minerals occur in soils. As a result of later work it was established 
definitely that clay materials in general are crystalline. Attempts 
have been made to determine the size of plate-like crystals of clays 
by measurement of the limiting velocity of settling in water; but 
the formulas used, which were deduced from the formula appli- 
cable to small spheres (using Stokes' law), are not very convinc- 
ing. On the other hand, x-ray diffraction and electron diffraction 
have been used to determine the crystal size and habit; these inves- 
tigations of the fine minutiae of the crystal structure seldom led 
to any definitely acceptable interpretations. 

With the advent of the electron microscope, it was soon 
realized that here is an instrument which is "particularly well 
adapted to the study of clay minerals; first, in order to charac- 
terize them by their appearance and, second, to assess, more 
exactly than has hitherto been possible, the contribution of their 
crystal habit to other physical and chemical properties." (Mar- 
shall, Humbert, Shaw and Caldwell) . 

Micrographs of many forms of clay have been taken in the 
Research Laboratory of the American Cyanamid Company 
(Stamford, Conn.) and at the University of Toronto, but prob- 
ably the most systematic investigations have been carried out in 
the Radiation Laboratory of the Ohio State University using the 
electron microscope built and operated under the direction of 
Professor A. F. Prebus. These investigations were carried on in 
co-operation with the Departments of Agronomy of the Ohio 
Agricultural Experimental Station (Wooster, Ohio), and of the 
Ohio State University (Columbus, O.), and with the Department 
of Soils of the University of Missouri. 

As a result of these studies definite details of the structure of 
crystals in many commercially important clays have been deter- 
mined. They report: 



Courtesy A. Prebus, Ohio State University 

Chrysotile asbestos (X 40,000). 

Courtesy American Cyanamid Co. 

Asbestos Fibers (X 21,000). 


(1) Montmorillonite (San Diego, Calif.), illite (from Goose 
Lake, 111., known commercially as grundite), beidillite (from 
Putnam silt loam, Missouri) and bentonite (from Rock River, 
Wyo.) consist of extremely thin plates, which have a very large 
specific area. In the first-named clay the plates are only about 
Imp thick. 

(2) Kaolinite (an English China clay sample) and dickite 
(from Chihuahua, Mexico), also have plate-like crystals, but 
much thicker than those of the preceding classes. 

(3) Nontronite (from Sandy Ridge, N. C.) and halloysite 
(from Maiden, N. C.) have lath-shaped crystals. 

(4) Attapulgite (from Attapulgus, Ga.) and magnesium ben- 
tonite (from Hector, Calif.) are very fibrous. This also is the 
structure of asbestos in powdered asbestos. (Page 50.) The 
above workers conclude that fibrous or lathlike clay minerals 
are much commoner than formerly supposed. 

"Highly characteristic features in the crystal habit of the 
different minerals are associated with their chemical and physical 
properties. Wide variations in particle shape can be closely 
correlated with extreme differences in the behavior of clays. The 
structures exposed by electron microscope studies are of more 
value for purposes of correlation of the physical properties of 
unfired clay. With proper interpretation, however, electron 
photomicrographs will also provide a satisfactory understanding 
of such properties as plasticity, hydration, green strength, and 

"Plasticity results from (1) the interplay of the attractive 
force tending to hold the clay minerals together, (2) the thickness 
of the water film between the plates, and (3) the lubricating 
properties of the water film. Best working properties depend on 
the relative thickness of the water film to the strength of the force. 
As plasticity changes with structure variations, electron photo- 
micrographs offer opportunity for valuable deductions from par- 
ticle shape. In montmorillonite, the extreme thinness of plates 
and consequently the great number of water films indicate that 
the order of plasticity will be high. The extreme amount of 
surface area is also closely correlated with a high bonding power 


and a high drying shrinkage. The structures of illite reveal quan- 
tities of surface intermediate in magnitude. 

"The hexagonal plates of kaolinite are stable. Their low base- 
exchange capacity is attributed to their stability. . . . The rela- 
tively low amount of surface per unit mass explains the low values 
for bond strength, drying shrinkage, and general plastic proper- 

"On hydration, a film of water envelops the exposed surface 
of the plates. Variations in amount and character of surfaces as 
exposed by the electron photomicrographs may explain the wide 
variations in swelling; low chemical stability of magnesium ben- 
tonite is closely correlated with the fibrous structure of this clay 


Mine Dust. Drilling and blasting operations produce silica 
dust in the air of a mine. If this dust is breathed into the lungs 
over long periods of time the condition may develop into sili- 
cosis, which is a serious disease among miners. The drilling dust 
appears to be more harmful in its effect than the blasting dust, 
but no significant differences between the two could be discovered 
by ordinary methods. The electron microscope was used to 
examine the dusts and to determine whether any important differ- 
ences did exist. 

Meteorologists and public health officers have long been in- 
terested in the measurement of the size distribution of dust par- 
ticles in normal and abnormal atmospheres. Various methods of 
capturing a sample of the dust have been tried; the earlier work 
depended oil bubbling the dust-laden air through water and let- 
ting the dust settle on a microscope slide at the bottom of the liquid 
column, or on blowing a blast of dust-laden air against a glass 
plate to which the dust particles tend to cling. 

Those interested in mine dusts in England developed a far 
more satisfactory collecting apparatus, the thermal precipitator, 
(Fig. 123.) The dust-laden air is drawn through the chamber 
past a thin metal wire which is heated electrically. For some 
reason, which is not entirely clear, the dust particles are repelled 
by the warm wire and deposited along the walls in quite a definite 
line.' A small microscope slide (or the object holder of an 









/y'/m on /nes/j 

Fig. 123. Plan view of thermal precipitator. 

electron microscope) is inserted in the position shown so as to have 
the dust particles deposited for observation. This apparatus 
clears the dust out of the air quite efficiently. With this thermal 
precipitator the dust sample can be taken right in the mine. 

A series of electron micrographs are taken over the width of 
the thermal precipitator deposit, for several positions along its 
length. From these photographs, projected on a screen at 
X 50,000 (5.0 cm = 1 j*), the particle sizes can be measured easily 
and the numbers in any particular size range can be counted. 
From these experiments it was concluded that the percentage 
of particles with diameters less than 0.20 p was much greater in 
the drilling than in the blasting dust. About 50 per cent of the 
particles in the drilling dust were less than 0.10 fi in diameter; about 
12 per cent of the particles from the blasting dust were of the 
smaller sizes. Optical microscope counts of samples taken on glass 
slides at the same time as the electron microscope samples were 
obtained gave good correlation in the range 0.10 to 2.0/*. The 
distributions of the dusts were very asymmetrical. 

The use of the thermal precipitator as a piece of standard 
equipment for mounting aerosol samples for electronic investi- 
gation is an important mounting technique, especially when a 
good distribution of the particles under the conditions in which 
they normally occur in the air is desired. 


Courtesy University of Toronto 

Mine dust (blasting) X 14,000. 

University of Toronto 

Mine dust (drilling) 
(X 14,000). 


,:.-..:> ..;- ^ *:' ; 'l... " 


Crystalline and Non-Crystalline Smoke Particles. The study 
of fine particles has a very important scientific interest entirely 
apart from commercial considerations. Plates on pages 120 to 
126 show the contrast between the appearance of the small par- 
ticles in the smokes consisting of the oxides of aluminum, 
magnesium, tungsten and zinc. In the case of aluminum, the 
particles are manifestly all spherical, while in all the others the 
structures of all the particles are definitely crystalline. In the 
nature of the case, the particles must have grown from the 
aggregations of molecules of the oxides; apparently here we see 
evidence of the struggle between the forces of crystallization 
and the forces of surface tension in the fashioning of the final 

In the pictures just mentioned the particles were deposited on 
a thin film (of collodion or formvar) but another technique 
used at Toronto has produced interesting results. Air laden with 
fine particles is drawn past a fine fiber (about one micron in 
diameter) stretched across the mesh holding the object. The 
particles stick to the fiber, as well as to one another, and an 
irregular layer of particles forms on the fibers. On pages 232 
and 233 are shown pictures (1) of carbon particles and (2) 
of particles of sodium chloride resulting from atomizing a salt 
solution into air. In the case of the salt, the crystals have been 
built up by tracing an enlarged projection of the negative thrown 
on the wall. It is an interesting study to try to imagine how 
these crystals are held together in such an apparently precarious 
fashion. See also lens paper, page 98. 

An additional picture is shown of carbon particles collected 
on fibers of three different diameters; it is quite noticeable that 
the smaller the fiber the greater the number of particles attached 
per unit of length. In each case the air sample was the same and 
the rate at which the air was drawing past the fiber was the same. 

Formation of Silver Deposits in Photographic Emulsions 

Investigations on the formation of the silver deposits in photo- 
graphic emulsions were carried out in the research laboratory of 
the Eastman Kodak Company with the help of an electron micro- 


scope built in that laboratory. A selection of the electron micro- 
graphs are reproduced here, with comments taken from a paper 
by C. E. Hall and A. L. Shoen. These results have revolutionized 
the theory of the photographic development processes. 

One of these plates, A (X 21,500), is the picture of a single 
silver bromide crystal exposed to an intense electron beam for 
some time. At first such crystals are opaque to electrons and are 
therefore quite uniformly black in such a picture taken with the 
electron microscope. On being exposed to an electron beam for 
some time, these crystals begin to show transparent holes and 
cracks as in this figure. It is thought that this appearance is due 
to the migration of silver ions within the lattice of the crystal. 
The appearance of the crystals gradually changes while the expo- 
sure to the electron beam is prolonged, but in time a stable state 
is reached. 

B ( X 30,000) is a picture of very small silver bromide crystals, 
known as Lippmann crystals: not developed or fixed. These 
would of course be quite invisible in the optical microscope. 

C ( X 25,000) is the picture of a silver bromide crystal exposed 
to an electron beam, as for the crystal A, and then fixed. This 
crystal was exposed to an electron beam, while it was observed 
in the electron microscope. It was then taken from the electron 
microscope and washed in sodium thiosulphate solution. This 
dissolved the silver bromide away but left the silver, shown in 
black. It is to be noted that the reduced silver does not cover 
the cross-sectional view of the crystal uniformly; the free silver 
gathers about certain preferred points. 

D (X 25,000) shows a group of silver bromide crystals ex- 
posed to an intense light beam, then put into the developer for 
a very short time, but not fixed. Some of the silver bromide 
crystals, the solid black portions of the picture, have not been 
affected appreciably by the developer. The thread-like struc- 
tures are filaments of silver, which have been reduced from silver 
bromide by the developer. 

E (X 25,000) shows crystals exposed to an intense light beam, 
then developed longer than those in D, but still only partially 




t . f 


% * 

| " W 

,# # 



. . '■'■.'.-■;■■■■ 


- ' . ■'-.■ ; : --. ■:..■:■ : 




Key to illustration on opposite page 

A. Silver bromide crystal exposed to intense electron beam (x 21,500). 

B. Undeveloped Lippmann crystals (x 30,000). 

C. Silver bromide crystal exposed to electron beam. Fixed (X 25,000). 

D. Silver bromide crystals. Partially developed. Not fixed (X 25,000), 

E. Partially developed silver bromide crystal. Fixed (X 25,000). 

F. A silver grain developed in amidol (X 25,000). 

G. Undeveloped Lippmann crystals (x 30,000). 
H. Developed Lippmann crystals (x 30,000). 


developed and then fixed. A faint outline can be seen in the 
background where the original crystal has left its trace in the 
gelatin. There are many small particles of silver present, con- 
firming the view that development starts at certain preferred cen- 
ters, presumably at the latent-image centers. 

F (X 25,000) is the picture of a silver bromide grain which 
had been exposed to intense light and then developed on the elec- 
tron microscope holder. Although this shows the kind of a silver 
deposit of which a developed silver grain is composed, it is not 
the same in general appearance as the grains developed on a pho- 
tographic plate. In the latter the filaments are so closely packed 
that most of the grain is completely opaque with filamentary 
structure in evidence only around the edges. 

This particular crystal was developed in amidol. With the 
exception of pure paraphenylenediamine, all of the common 
chemical developers produce this typical filamentary structure 
with some secondary differences. Metol and amidol, for example, 
produce very fine filaments, whereas hydroquinone produces 
coarse filaments. The extent to which the filaments can wander 
away from the original crystal is influenced by the rigidity of the 
surrounding gelatin. 

G (X 30,000) shows Lippmann crystals exposed to light and 
partially developed but not fixed. The round dots are the silver 
bromide crystals: it will be noticed that each one has a projection 

H ( X 30,000) shows Lippmann crystals exposed, fully devel- 
oped and fixed. Each small crystal has grown into a single fila- 
ment of silver which is much longer and thinner than the original 

Molecules of Organic Materials 

In the discussion of the properties of such solutions as pro- 
teins, colloid scientists have been driven to assign enormous values 
to molecular weights, and, consequently, a large size to such 
molecules. In fact, these scientists have great difficulty in agree- 
ing on the definition of a molecule. The accompanying pic- 







• V- ■■ . 

■ .. 

%■:■ ''MmwSmMk.. 



v V :S #illP 

** , % 

,*»If r 


;.'■':' * > 


tifllfs^ : 


■■'•. " ' ■ 

. : *%fc''ifci. 


«■ "« 







Courtesy R.C.A. Laboratories 

Polymerized vinylchloride (x 210,000) 

Courtesy University of Toronto 

Cuprene, a polymer of acetylene found in the presence of copper ( X 22,700) 


tures show various organic materials, but it is very hard to tell 
just what bearing these pictures have on our ideas of molecules 
of these substances. 

Electron Micrographs of Fibrous Proteins.* "Structures re- 
flecting the molecular organization of protein fibrils have been 
demonstrated with the electron microscope at the Massachusetts 
Institute of Technology. Fibrils of collagen, which is the chief 
protein constituent of tendon, skin and connective tissue gener- 
ally, show a regular banded structure, as shown on page 269. The 
average value of the spacing (the distance between centers of 
consecutive like bands) is about 640 A which is in agreement 
with x-ray diffraction measurements made by R. S. Bear. This 
banded structure is characteristic of collagen fibrils from different 
parts of the body and from different animals, and results from 
an alternation in density of the protein along the fiber axis. Finer 
detail within the fundamental period has been revealed by the 
use of heavy metal stains. The stains are preferentially adsorbed 
to produce at least five stained sub-bands within the fundamental 
period. (A is another symbol for A.U.) 

"Detailed structural patterns have also been demonstrated in 
protein fibrils from molluscan muscles by the same group. In 
this investigation the fibrils were treated with selective electron 
stains of high mass and molecular weight in order to improve 
the contrast in the electron image. An electron micrograph of 
typical fibrils from the adductor muscle of the clam, Venus mer- 
cenaria, is shown here also. In this case the electron stain was 
phosphotungstic acid. The stain combines with the protein at 
regular intervals along the fiber axis as a result of a periodicity 
in the protein structure. The average spacing as determined 
from some 200 fibrils was found to be 145 A and the maximum 
deviation from this value was about 10 per cent. 

"If the stain is applied in suitable amount, each cross band 
can be seen to consist of a series of equally spaced spots, as shown 
herewith. The fibrils are ribbon-like and usually dry during 
preparation with a flat side against the supporting film. Stained 
spots in consecutive cross bands are aligned at definite angles to 

* Communicated by C. E. Hall. 






Electron micrograph of collagen fibers from rat tail tendon (X 37,500) 




Portion of fibrils from adductor muscle of Venus mercenaria stained with 
phosphotungstic acid, showing stained cross bands (X 110,000). 

Courtesy C. E. Hall, Massachusetts Institute of Technology 

Fibril from adductor muscle of Venus mercenaria showing two-dimensional 
lattice (X 236,000). 


the fiber axis as indicated by the angles « and fi drawn on the 
electron micrograph. There is a definite relation between the 
angles a and /? so that tan « = f tan £. The disposition of spots 
in the cross bands and diagonals forms a remarkably perfect 
geometrical pattern which is shown diagrammatically in the plate. 
The pattern is laterally asymmetrical. If through any spot a 
line is drawn parallel to the fiber axis, it passes through a series 
of spots spaced five bands apart. The length of the unit cell is 
therefore 5 X 144 = 720 A, as indicated in the plate. The lateral 
distance between spots is constant and is equal to about 193 A. 

"Earlier observations on the same material stained with osmic 
acid showed cross bands which were 360 A apart in the fiber 
axis direction. Apparently in the case of osmic acid, the stain 
reacts with groups spaced one-half the fundamental axis period 

"Concurrent with the electron microscope study R. S. Bear 
has investigated the small angle x-ray diffractions from the same 
muscles. The x-ray pattern consists of a series of diffractions 
on the meridian and two row lines, one on either side of the 
meridian. From the layer line positions Bear concluded that the 
fundamental fiber axis period is 725 A. The diffraction orders on 
the meridian are strong and their indices are multiples of five out 
to the 40th. This indicates a sub-period equal to one-fifth the 
fundamental period, or 145 A. Thus the dimension of the funda- 
mental period and the sub-period as obtained from x-ray diffrac- 
tion data are in perfect agreement with those obtained from the 
electron micrographs. 

"The distance between the row lines in the x-ray diagram 
indicates a lateral spacing of 325 A, which does not correspond 
within experimental error to any distance measured in the elec- 
tron micrographs. This lateral distance is undoubtedly related 
to the 193 A lateral distance in the plate. A clarification of this 
aspect of the analysis is expected from further studies now in 
progress. In general the x-ray and electron microscope results 
are in remarkably good agreement; the discrepancy in the lateral 
period serves to emphasize the value of obtaining both types of 
information in a structural analysis of this nature." 


Application to Medicine 

The electron microscope is undoubtedly destined to play an 
important role in medicine. Already a few research groups have 
undertaken systematic investigation of various medical problems. 
Naturally this work cannot be speeded up in its preliminary 
stages because the electron microscope reveals so many details 
which are entirely new that many pictures have to be taken under 
varying circumstances before even the most experienced worker 
can attempt to give a convincing interpretation. In the early 
stages of the application, pictures were taken of disease agents, 
many of which are below the range of optical microscopes. Even 
in the case of those which are visible in ordinary microscopes 
many new details of structure are made visible. It is not beyond 
the realm of probability that we may soon be able to make con- 
vincing studies of the life history of these minute individuals 
which serve such an important function in the human body, for 
good or ill. 

The earliest systematic work in this field in America was that 
undertaken at the RCA Laboratories, # first under L. Marton, 
and later, under James Hillier, in cooperation with W. M. Stanley 
and T. F. Anderson of the Rockefeller Institute for Medical 
Research and Stuart Mudd of the School of Medicine of the 
University of Pennsylvania. Recently the two latter investigators 
have been cooperating in Philadelphia, as Anderson is attached 
to the Johnson Foundation for Medical Physics at the University 
of Pennsylvania. Electron micrographs of a few bacteria were 
taken at Toronto when the work was begun there, but during 
the war years systematic medical investigations have yielded 
precedence to problems relating directly to the war. 

"New possibilities for the study of the structure of bacteria, 
rickettsias, viruses, phage and of fine structure in tissue have been 
opened up by the electron microscope" (Mudd and Anderson). 

*A£ter the first few pictures of bacteria were taken by Dr. L. Marton at the RCA 
Laboratories, the RCA furnished the funds to the National Research Council for a 
Committee on Biological Applications of the Electron Microscope. This comprised 
about 17 specialists who worked according to a pre-arranged calendar at the RCA 
Laboratories with Dr. T. F. Anderson, the RCA Fellow on the Electron Microscope, 
National Research Council, 1940-42. 



%,, IP 

A0 mm 9K% 

• •* 


: ••♦ 



■ """■■ 



Tobacco mosaic (old preparation) (X 50,000). 









Courtesy RCA Laboratories 

Tobacco mosaic (fresh preparation) (X 50,000). 


Already enough progress has been made in this field to enable 
us to classify the lines of future progress that have opened up. 
These may be listed as follows: 

(a) The morphology of disease agents of various types. 

(b) The interaction between disease agents and antibodies. 

(c) The life history of a bacterium. 

(d) The seat of biological chemical reactions induced by 

or in bacteria. 

(e) Microtome sections. 

There is great promise that, in time, sections may be cut thin 
enough by a microtome to be used as objects in the electron 
microscope. Special methods of staining also indicate future 
usefulness in this field. 


Size and Shape. The first new feature is that the electron 
microscope reveals the size and shape of individual disease agents 
not resolvable by the optical microscope. Stanley and Anderson 
have made a very complete study of the tobacco masaic virus. 
They point out that the limits of the small entity known as a virus 
have been sought by various methods — ultrafiltration through 
collodion membranes, microphotography with ultraviolet light 
using quartz or fluorite lenses, fluorescent microscopy, and special 
staining techniques — but for the very small specimens of virus the 
judgment of size (and shape) is dependent on more indirect 
evidence, such as rate of sedimentation (in the ultracentrif uge) , 
diffusion, double refraction of a flowing stream containing the 
virus in suspension, viscosity of suspensions and x-rays. Most of 
the latter methods had been developed for work on colloidal 

As a result of these various methods the tobacco mosaic virus 
was considered to be of a rod-like form about 12 m/x in diameter 
and 400 m/x long; a molecular weight of 40X10 6 was deduced, 
also from these experiments. Stanley and Anderson found from 
electron micrographs that the predominant unit in the suspen- 
sion was rod-like, with a diameter of 15 m/x and a length of 280 





Cottrte.ry .Dr. Stuart Mudd 

Bacillus anthracis (X 4,800). 

This is very good evidence of a continuity in measurement 
from the range of the ordinary microscope down to that of the 
electron microscope; it also indicates that the virus has not suffered 
any great change in being mounted in the vacuum chamber of 
the electron microscope. 

Stanley and Anderson state as their considered opinion: 
"Although excellent micrographs of bacteria have been obtained 
by means of this apparatus and have proved of value in supple- 
menting information already available, it would appear that the 
electron microscope will be of greatest value in the microscopy 
of objects having sizes between 5 and 250 mju,, a range not covered 
by the light microscope and one in which practically all viruses 


have been found to fall. The electron microscope offers the 
possibility of securing micrographs of individual virus particles 
and thus of establishing their sizes and shapes with some precision. 
It should also be possible to determine the extent of the variation 
in the size and shape of a given virus and even perhaps learn some- 
thing of the mechanism by means of which a virus particle is 
duplicated within the host and of the nature of the difference 
between strains of a given virus." 

The Cell Wall and Flagella. Dr. Stuart Mudd and his co- 
workers have laid considerable stress on what electron micro- 
graphs show with regard to the cell wall. Their general con- 
clusion is that bacteria possess a definite cell wall of substantial 
structure; this wall is in a sense independent of the protoplasmic 
contents which seem to be contained in a separate protoplasmic 
membrane. As a result of observation on ^-hemolytic streptococci 
(Lancefield's Group A) they state that "taken together, the 
phenomena shown would seem very definitely to indicate that 
streptococcal cells do possess rigid outer membranes whose con- 
tinuity accounts for the grouping in chains and which are dif- 
ferentiated from the inner protoplasm." Again as a result of 
photographs of Bacillus subtilis, B. megatherium and B.anthracis, 
they state that "these possess walls that are definite, solid mor- 
phological structures. An inner protoplasm may shrink away 
from this wall, or escape following injury, leaving the cell wall as 
a 'ghost,' which has essentially the shape of the intact cell. The 
flagella of the B.subtilis are continuous with the cell wall: the 
spores appear to be very dense, rigid bodies." Such micrographs 
as those of the pneumococcus (taken at Toronto), show very 
clearly the existence of a thick capsule about the central cell. 

Structure in the Protoplasmic Contents. Micrographs of the 
tubercule bacillus taken at Toronto and other places give indi- 
cation of dense spherical bodies in the protoplasmic contents of 
the cells. On page 286 are two micrographs of single cells of 
M.tuberculosis from the same medium, one perceptibly smaller 
than the other. In the larger bacillus four very distinct granular 
bodies are present, while in the smaller there are four correspond- 
ing centres of cloudiness, as though denser granular bodies were 



Courtesy University of Toronto 

Pneumococcus (X 67,000) taken from the peritoneal cavity of a dead mouse. 


Courtesy University of Toronto 

Pus cell with pneumococcus. 

in process of formation. These may be examples of young and 
old germs. 

Dr. Mudd and his colleagues have taken micrographs of the 
cells of Mycobacterium tuberculosis (hominis) and find that 
the cell wall appears to be very delicate. There were many 
granules in the field, a large number apparently adhering to the 
cell wall. "The thirteen largest of the black granules seen within 
the protoplasm range from 70 to 230 m//, in diameter." These 
granules have been recorded many times previously but no satis- 
factory interpretation has as yet been established. 

The plate on page 238 "shows cells of Vibrio schuylkiliensis. 
Definite circumscribed granules are again seen within the proto- 
plasm. The nineteen most definite of these range in diameter 
from 90 m^ to 230 m/i. In addition to having discrete granules, 
the whole bacterial protoplasm appears to be either finely or 
relatively coarsely granular; this appearance is in our experience 
rare and may be a coagulation artefact due to drying or to the 
electronic bombardment. The protoplasm is obviously shrunken 
away from the relatively 'transparent' cell wall in various cells." 

Dr. Mudd concludes that "granules in bacterial protoplasm 
have been variously interpreted as nuclei or nuclear equivalents, 
reserve food material, reproductive elements, or otherwise, with- 


Courtesy Dr. Stuart Mudt 

Corynebacterium diphtheriae and Tellurium (x 29,700) 


out any of these interpretations having become established. The 
electron microscope shows fine structure within bacteria with 
clarity and detail not hitherto possible, and, when coordinated 
with cultural and cytologic procedures should contribute to 
eventual understanding of these structures." 

It cannot be emphasized too strongly that electron micro- 
graphs alone will not solve everything about these small struc- 
tures. Cooperative research in all bacteriological fields will be 


As an example of similar morphological results in the coccus 
type of bacteria we may quote from the investigation of Knaysi 
and Mudd. "The cell of Staphylococcus flavocyaneus (page 250) 
contains one or more granules which have solubilities similar to 
nucleo-proteins and which often appear constricted or in pairs. 
In very young, actively growing cells, the granules demonstrable 
at high voltages are reduced in size and there is evidence that the 
nuclear material is then partially in solution. A strain of Neisseria 
meningitidis also shows granules which are insoluble in hot water 
and which are likely to be of nuclear nature. 

"On the other hand, the cells of strains of Neisseria gonor- 
rhoeae, Staphylococcus aureus, and Streptococcus pyogenes^ 
strains 1048 M and C203M, appear homogeneous at all voltages" 
{i.e., show no contrast even with electrons of high velocity). 
Drs. Knaysi and Mudd here conclude that the results with the 
coccus cells "support the view that different bacteria may contain 
nuclear material in different states, and that the state of the nu- 
clear material may change with the development of the cell." 

Dr. Anderson and investigators at the Hospital of the Rocke- 
feller Institute for Medical Research, New York, found "remark- 
able regularity in the morphology of the elementary body of 
vaccinia. The virus bodies apparently have internal structure 
and some sort of limiting membrane." Plates on page 280 show a 
micrograph of small-pox virus taken at Toronto. 

Finally we may refer to the investigations of this group on 
bacteria of spirochete type. They reach the following conclu- 





Smallpox virus (X 113,400) 

Courtesy University of Toronto 

Smallpox viri (X 36,000). The barrel shape may be due to centrifuge action; 
structure visible may or may not be actual. 


"Electron micrographs of the Nichols-Hough, Kroo and 
Reiter cultured strains of Treponema pallidum, of treponemes 
of the virulent Nichols-Hough strain from a rabbit syphiloma, 
and of cultured strains of Treponema macrodentium and Tre- 
ponema microdentium are presented and the morphology of the 
treponemal cells described. 

"A delicate cell wall or periplast encloses the inner proto- 
plasm of treponemata; this periplast may connect adjoining cells 
until transverse cell division is completed; thereafter it may 
extend beyond the cell protoplasm as a terminal filament. No 
evidence of a differentiated axial filament within the protoplasm 

is found. 

"Flagella, often in groups of four, are found along the sides 
or near the ends of the cells of T. pallidium and T. macrodentium. 

"Dense granules, 40 to 90 m/x in diameter are often found 
within the spirochetal protoplasm. 

"Irregularly spheroidal, dense bodies, 150 to 500 my. in 
diameter, are often found attached to the spirochetal cell, fre- 
quently near the end; such a dense body may be in close apposi- 
tion to the outside of the spirochetal cell wall or may be con- 
nected to it by a short stalk. The evidence concerning these 
bodies seems to support the interpretation that they are asexual 
reproductive bodies." 

Interaction between Bacteria and Antibodies 

Electron micrographs have been used to give evidence of the 
nature of the interaction between bacteria and their environ- 
ment: two examples of this will be given: (1) the action of 
antiserum on tobacco mosaic virus and (2) the action of bacterio- 

Anderson and Stanley (in a cooperative investigation with 
the RCA Laboratories and the Department of Animal and Plant 
Pathology of the Rockefeller Institute for Medical Research, 
Princeton) have investigated the action of various substances on 
tobacco mosaic virus and report the following: 

"Electron micrographs of tobacco mosaic virus deposited on 
a collodion film show that the molecules are about 280 m/x, long 
and about 15 m/*, wide. 





Courtesy University of Toronto 
Rickettsia (typhus) (X 13,000). Apparently crystals are formed at edges of bacillus. 




Courtesy Dr. T. F. Anderson 

Rickettsia-marine typhus (X 16,000). 



"Micrographs of a mixture of virus and normal rabbit serum 
show virus particles of normal size and indicate little or no absorp- 
tion of particles from normal serum onto the virus molecules. 
Similar results were obtained with mixtures of tobacco mosaic 
virus with antisera to tomato bushy stunt, potato latent mosaic, 
and tobacco ring spot viruses. 

"A mixture of tobacco mosaic virus and tobacco mosaic 
virus antiserum from rabbits, when dried on a collodion film 
an hour after mixing and examined by means of the electron 
microscope, shows particles about 60 m^ wide, about 300 m/x 
long, and having fuzzy profiles. The increase in particle width 
and the fuzzy appearance are regarded as indicating that the ends 
of asymmetrically shaped molecules from the serum react specifi- 
cally with the antigen molecules. No reaction between anti- 
tobacco mosaic virus serum and bushy stunt virus was demon- 

"When the mixture of antigen and antiserum is applied to 
a collodion film several hours after mixing, an irregular frame- 
work of thickened antigen molecules may be seen. It is this 
framework which makes up the antigen-antiserum precipitate. 
The results demonstrate the usefulness of the electron micro- 
scope and of a large and distinctively shaped antigen such as 
tobacco mosaic virus in the study of the antigen-antibody 


Luria and Anderson (in a cooperative investigation with the 
Bacteriological Research Laboratories of the Department of Sur- 
gery of Columbia University and the Research Laboratories of 
the Radio Corporation of America at Camden) have investigated 
the action of bacteriophages. 

"Bacteriophages, or bacterial viruses, are a group of viruses 
reproducing in the presence of living bacterial cells. Bacterio- 
phages are particulate, and convincing evidence exists that (1) 
one particle of phage is sufficient to originate the lysis of a 
bacterial cell; in the lysis, a variable number of new phage par- 
ticles (an average of 100 or more) are liberated per cell; (2) the 
elementary particles of each phage strain seem to have a charac- 
teristic particle size as determined by any one of various indirect 






^.'H r ' ■■ 

„W;,;I, y 


' : W- : - 




Bacillus Novy (X 34,200). 

Courtesy University of Michigan 



Courtesy University of Michigan 

Bacillus Novy (X 34,200). Continuation from end of previous pressure. 

methods of investigation (ultrafiltration, radiosensitivity, diffu- 
sion), and diameters ranging from 10 to 100 m/x have been 
obtained for the various strains, depending on the method of 
investigation, although diffusion experiments occasionally yield 
still smaller values." 

The results are reported as follows. With bacteriophage anti- 
coli PC the micrographs show "the constant presence of par- 
ticles of extremely constant and characteristic aspect. They 
consist of a round "head," and a much thinner "tail," which gives 
them a peculiar sperm-like appearance. The "head" is not homo- 
geneous but shows an internal structure consisting of a pattern 
of granules, distinguished by their higher electron scattering 
power. Deviations from the usual symmetrical internal pattern 
may be due to varying orientation of the particles or to other 
factors as yet unknown. The diameter of the head appears to 
be about 80 mju,; the tail is about 130 m/x long. 



l§S: ; 


Courtesy University of Toronto 

Tuberculosis bacillus (X 31,400). 

Courtesy University of Toronto 

Tuberculosis bacillus ( X 26,400) . 
These two are from the same culture. 


"When allowed to stand a few minutes in the presence of sen- 
sitive bacterial cells of Escherichia coli, PC, the particles described 
above are readily adsorbed. They appear to stick to the bacteria 
either by the head or by the tail. Other conditions remaining 
constant, the number of particles adsorbed on a bacterium in- 
creases with the time of contact, although it is difficult at the 
present time to differentiate between adsorption and reproduc- 
tion of the particles on the cell wall. By allowing the phage to 
stay in contact with bacteria for a time of the order of the mini- 
mum time of lysis (21 minutes for PC phage, Delbriick and 
Luria) it is possible to observe bacterial cells extensively damaged, 
surrounded by a very large number of particles, probably newly 

Similar observations were made using bacteriophage anti-coli 
P28 and B — anti-staphylococcus 3K. 

This is undoubtedly a very important field of research. 

The Life History of a Bacterium 

There is great hope that the electron microscope will enable 
us to trace the life history of various forms of bacteria, both in 
their undisturbed state and after they have been attacked by 
various drugs which have recently been applied so successfully. 
There is very little doubt that variations in the appearance of 
the protoplasmic contents of the cells and even the variations 
observed in the size of these cells will be found to depend on the 
life history of the organism. 

The Seat of Biological Chemical Reactions induced by or in 
Bacteria; Special Staining Technique 

We shall quote from a paper by Morton and Anderson 
(Department of Bacteriology, University of Pennsylvania, School 
of Medicine, Philadelphia and the RCA Laboratories, Camden, 


"Since the beginning of the science of bacteriology, inves- 
tigators have pursued the study of the chemical reactions brought 
about by bacteria. As a result there has accumulated a tremen- 
dous mass of knowledge concerning the types of reactions in- 


duced by bacteria, and the particular reactions which charac- 
terize various species. However, in all this work it has not been 
shown definitely whether the reactions in question take place 
exclusively on the cell surface, as some investigators maintain, or 
inside the cell, or both. 

"With a powerful new tool, the electron microscope, now 
available, it was thought desirable to attempt to determine the 
site of some chemical reaction brought about by a bacterial cell. 
Since the electron microscope detects variations in density, such 
a reaction should involve as products large particles or crystals 
resolvable with the microscope; i.e., the particles must have diam- 
eters of at least 50 A (5 m/x). Another possibility would be to 
select for study reactions which involve substances which are 
capable of adsorbing particles or substances such as large protein 
molecules, organic compounds, or inorganic salts, any of which 
would give rise to a region of high density within or upon the 
cell. The addition of material to a suspension would, however, 
give rise to the possibility of the production of artefacts. We 
have consequently chosen for initial study a reaction of the first 
type, one brought about by an organism in a medium on which 
it produces an insoluble substance. 

"The reaction chosen for study was discovered in 1900 by 
Klett. He observed that many microorganisms, including the 
diphtheria bacillus, when grown in or upon culture media con- 
taining tellurite or selenite salts, reduce them to the free metals. 
Conradi and Troch proposed a medium containing a tellurite 
salt for an aid in the diagnosis of C or yne bacterium diphtheriae, 
because on this medium the colonies of this organism have a 
characteristic black color due to the reduced tellurium, and 
because the medium exerts a selective bacteriostatic effect, allow- 
ing the diphtheria bacillus to grow but inhibiting certain other 
microorganisms. Although Klett observed gray particles inside 
and outside the cells of Bacillus anthracis grown on tellurite 
media, it is not known where the reaction occurs in the case of 
C. diphtheriae. 

"We have obtained micrographs with the electron micro- 
scope of cells from diphtherial colonies growing on blood ex- 


tract agar and on potassium tellurite chocolate) agar. The plate 
on Plate 278 shows unstained cells of C. diphtheriae from blood 
extract agar. The polar granules appear as black circles, while 
the remainder of the cell is light gray. Since, as previously men- 
tioned, electron micrographs reveal differences in densities, this 
is to be interpreted as indicating that the polar granules are spheres 
or possibly plates of high density, while the remainder of the 
cell is of a relatively low density, in the dry state. 

"Plates (not here) are of unstained cells grown on potassium 
tellurite chocolate agar. Inside the cells, and in addition to the 
polar granules, are seen minute needle-like crystals of high 
density. If these crystals are actually tellurium metal they should 
dissolve in the presence of an oxidizing agent such as bromine. 
To test this point, a drop of bromine water was added to 1 cc 
of a suspension of cells from black colonies on potassium tellurite 
chocolate agar. The black color of the mass of cells immediately 
disappeared: likewise, the needles were no longer observed in 
electron micrographs of cells so treated. Thus it appears highly 
probable that the black color is due to tellurium metal which 
exists, at least in part, in the needle-like forms shown in the 

"More important than the identification of the crystals is the 
location of the site of the reduction of the tellurite. In a few 
exceptional cases crystals actually penetrated from inside out- 
wards beyond the cell boundary, possibly during the drying of 
the preparation. In other cases the crystals distort the cell out- 
line without penetrating. In no cases were crystals observed to 
lie wholly outside the cells. Since the majority of the crystals 
are contained wholly within the cells, it is to be inferred that 
the tellurite or tellurous ion is able to diffuse through the cell 
wall and is there reduced to tellurium metal which is precipitated 
inside the cells (see Plate, page 278) . 

"Similar crystals have been observed within cells of C. 
xerosis grown on tellurite chocolate agar." 

Another application of tracing the chemical action is found 
in the development of special staining technique devised to show 
structural outlines more clearly. Mudd and Anderson report 


on the effects of heavy-metal salts on individual bacterial cells. 
The plates, pages 157 and 158, gives an example of this work. 

"The two pictures of Vibrio comma show the normal cells 
of the cholera vibrio and the cells after brief exposure to a lead 
salt. The last is particularly interesting because it illustrates the 
mechanism of injurious action. Evidently the salt diffuses 
through the cell wall, combines with the protoplasmic mem- 
brane, showing momentarily a dark line of lead compound; then 
the injured protoplasm begins to exude from the cell. 

"The physical basis of contrast and image formation in elec- 
tron micrography is considered in relation to the possibility of 
recording selective chemical effects on cell components. A 
technology of selective microchemical analysis, equivalent to 
differential staining, is suggested as practicable in electron microg- 

"Electron pictures of bacteria after exposure to salts of heavy 
metals have shown the bacterial inner protoplasm, but not the 
cell walls, to be selectively, darkened; shrinkage, coagulation, or 
escape of protoplasm from the injured cells may result and are 
recorded in the electron micrographs. 

"Recording of the action of germicidal agents on individual 
bacterial cells is indicated as one promising field of application 
of microchemical analysis with the aid of the electron micro- 

Microtome Sections 

The first question asked by the medical investigators inter- 
ested in the study of various abnormal growths is, "What is the 
prospect of using the electron microscope with thin sections 
such as made by a microtome?" The experimental difficulties 
are due to the fact that electron streams given by the ordinary 
accelerating fields (40 to 60 kilo volts) cannot penetrate sections 
as thick as the thinnest of those given by an ordinary microtome. 
If higher-velocity electrons are used they may penetrate every- 
thing and show little contrast. For 60 kv, sections should be 
less than 0.25 fi in thickness, preferably 0.1/x. 


Some progress has been made toward producing much thinner 
sections by A. G. Richards and co-workers (at the Zoological 
Laboratory, University of Pennsylvania). They have succeeded 
in producing sections 0.1/x thick and have published micrographs 
of striated muscle of the cockroach. 

Stereoscopic Photography with the Electron Microscope 

Everyone is more or less familiar with stereoptican views 
and their power to present to our vision a three-dimensional view 
of objects— in other words, their power to give the impression 
of depth to a picture. This, of course, presupposes that objects 
are in focus for our eyes for a considerable depth; any one 
object appears to be distinctly behind or in front of other objects 
in the picture. In everyday vision our eyes show the same power 
of "depth of focus"; one can look at an audience and see rows 
and rows of persons quite distinctly. 

The focussing power of the electron microscope resembles 
that of the camera and the eye: objects are in focus for a relatively 
great distance, one behind another. Consequently, stereoscopic 
pairs of pictures can be taken of specimens in three dimensions by 
this microscope and viewed by means of the stereoscope in the 
ordinary way. This procedure is impossible with the ordinary 
high-power light microscope because it has extremely small depth 
of focus. Relatively speaking, if our eyes had a small depth 
of focus, we should have only one row of persons in an audience 
in focus; persons in all the other rows would be out of focus and 
therefore blurred and indistinguishable. 

The making of stereoscopic pictures with the electron micro- 
scope is simple in conception, but requires considerable practice 
in accomplishment. After a picture is taken, the object is tilted 
through a small angle and a second picture of the same field is 
taken in the second position. These two pictures are used in the 
ordinary manner to give a stereoscopic view of the specimen. 
A typical pair of stereoscopic pictures is shown in the upper 
plate on page 232. These are pictures of carbon particles which 
have collected on a fine rubber fiber. Air laden with the carbon 



particles was blown gently past the fiber. It is interesting to 

note the manner in which the particles build up one upon another. 

A special object holder cartridge, designed by Hillier, is used 

Fig. 124. The special object holder used for securing stereoscopic photographs 
in the electron microscope. 

Position For Firvt 

Position For Second 

Fig. 125. Schematic drawings used to explain the use of the special object holder 
when obtaining a stereoscopic view with the electron microscope. 

to secure these photographs, Fig. 124. The cartridge is com- 
posed of two cylinders, with the axis of the small one at an angle 
of 6° to that of the large one. The axis of the large cylinder is 
parallel to that of the microscope when the picture is being taken. 



Courtesy Dow Chemical Co. 

Crystallographic etch attack in a Mg-Al alloy. Polystyrene-silica replica (X 4,500) 

Courtesy Dow Chemical Co. 

Twin boundary after crystallographic etch attack in Mg-Al alloy. Polystyrene- 
silica replica. (X 4,500). 



Courtesy Dozv Chemical Co. 
Surface of an etched quartz crystal. Calystyrene-silica replica (X 6,900), 

The object holder, which is a mesh of fine wire, is situated under 
a small metal cap at the end of the cartridge. 

The working of the holder in taking stereoscopic views is 
shown in Fig. 125. Let it be assumed that a first photograph 
has been taken with the two cylinders marked A and B in the 
position indicated by the dots. After the first exposure of a 
chosen field has been taken, the holder is removed from the micro- 
scope and B is rotated through 180° to its intermediate position 
relative to A. The cartridge is then rotated through 180° as a 
unit until the cylinders occupy the final positions as shown in the 
third figure. The holder is returned to the microscope in this 
position and a second photograph is taken of the same field. By 
this procedure the object is tilted through 12° and the relative 
position of the object with respect to the beam is unaltered. 
Twelve degrees is the approximate angle subtended by the dis- 
tance between normal human eyes at a distance of 25 cm from 
the eyes. This method of tilting the object by two rotations is 
necessary because it is unsatisfactory, mechanically, to attempt to 


tilt the object directly, although such a direct tilting process 
would serve just as well if it could be done easily and accurately. 
The hole in the lower end of the cylinder B has a diameter of 
0.5 mm and it leaves 12 openings in the mesh uncovered to the 
electron beam. The operator can easily examine the 12 openings 
and with the aid of a map of the field can rediscover the particular 
field to be photographed the second time. 

Surface Replicas 

When microscopists came to the problem of obtaining a 
highly magnified view of the surface of an opaque substance, 
such as a metal, they had to develop a new method of illuminating 
the surface from above. In this way microphotography was ap- 
plied to the study of etched surfaces of pure metals and alloys. 
The limitation as to resolving power due to the wave length of the 
light used was, of course, present also in this case. With the 
development of the electron microscope the question soon sug- 
gested itself, "Can this new instrument be applied to the surfaces 
of opaque materials?" 

Taking the cue from microphotographic practice, various 
methods of trying to learn something of the detail of these sur- 
faces by making use of electrons reflected from them have been 
attempted. The difficulties of interpretation of the results, in- 
herent in all these methods, led to the suggestion of the use of 
replicas, i.e., thin films to which the fine markings or defects of 
the metal surface can be transferred. For example, a solution of 
formvar may be poured on the surface so as to cover it with a 
thin skin when the solvent evaporates. The formvar penetrates 
all surface crevices and when the film sets it can be pulled off 
the metal surface. The film will be plain on one side and on the 
other will bear the contours which are the "negative" of those on 
the metal surface. The thickness of the film is such that the elec- 
trons penetrate the film, but differences in thickness would bring 
about differences in the blackening of the photographic plate. 

The modern study of thin films goes back to the fundamental 
work of Langmuir at the General Electric Research Laboratory 
(Schenectady) ; Dr. Schaef er of this laboratory is a pioneer in the 


production of surface replicas. Earliest suggestions of the appli- 
cation to the electron microscope are those of Mahl (oxide films 
on aluminum surfaces), and Schaefer and Harker (plastic films, 
e.g., formvar). 

At first only one film was produced. No matter how the film 
was applied to the surface, it had to be removed. This was done 
in various ways, e.g., by stripping the film from the surface with 
the aid of a razor blade to pry it up at the start, or by dissolving 
away the original specimen in some solvent which did not attack 
the material of the film. Such a replica of course gave a surface 
which was a negative of the original surface. 

It soon developed that positives provided a more satisfying 
object to photograph. These positives have been produced in 
several ways, two of which will be described. 

(1) That of Zworykin and Ramberg (RCA) utilizes a 
negative of silver and a positive of collodion. "A thick layer of 
metal, e.g., silver, is evaporated in a vacuum chamber onto the 
surface to be studied. This metal film is stripped mechanically 
from the surface, yielding a negative metal replica of the surface. 
If too thin for the stripping process, the metal film may be rein- 
forced by electroplating. Next, a dilute solution of the final 
replica material, e.g., a 1 per cent solution of collodion in amyl 
acetate, is flowed over the negative replica surface and is per- 
mitted to dry. Then the metal replica with the adhering thin 
film is immersed for several hours in a suitable solvent for the 
metal — 2 to 3 normal nitric acid in the case of silver. At the end 
of this period the metal is completely dissolved and, after washing 
in distilled water, the residual positive replica film may be placed 
on the fine-mesh screen (e.g., 250-mesh copper or stainless steel) 
which acts as the object support in the electron microscope." 

(2) That of Heidenreich and Peck (Dow Chemical Com- 
pany) utilizes instead a negative of polystyrene, and a positive of 
silica. "The technique developed is a two-step process in which 
the first replica is of molded polystyrene and the second, thin — 
film replica is evaporated silica. The formation of this molded 
plastic replica is simple in principle, consisting of simply making 
and impression of the original surface in a material which can be 


readily handled in the remainder of the process. The second 
replica which is observed in the microscope was discovered by 
chance since the property which makes it suitable for this purpose 
was not predicted. This property is the high mobility of silica 
condensing from the vapor both upon polystyrene and upon the 
silica in its vitreous state. The silica films are themselves struc- 
tureless to the electron microscope and electron diffraction shows 
them to be amorphous. They are easily removed from the 
styrene by the use of ethyl bromide solvent and washed free of 
any residual styrene." 

The above workers found that the silica replicas gave very 
much clearer and more highly resolved pictures of surface in- 
equalities than did such substances as formvar. 


In any account of the state of progress in any scientific field 
where developments are so rapid as has been true of the electron 
microscope it is difficult to tell where to stop. Any stopping 
place is sure to be abrupt and every reader may be inclined to 
ask of an author, "Why did you stop there?" 

A large number of these instruments are now in use and a 
new science of electron optics has come into being. One very 
substantial sign of the permanence of this development is the 
organization of a very active group of electron microscope 
specialists as the Electron Microscope Society of America. 

This society was organized in November, 1942, as an Asso- 
ciate Society of both the American Institute of Physics and the 
American Association for the Advancement of Science. It was 
organized on the recommendation of a committee of three — Pro- 
fessor G. L. Clark (Illinois), Professor O. S. DufTendack (Michi- 
gan) and Dr. L. A. Matheson (Dow Chemical Company). 

The first official meeting of the new society was held at 
Columbia University, January, 1944, and a second meeting in 
Chicago in the autumn of the same year. The Secretary- 
Treasurer of the Society is Dr. M. C. Banca, RCA Mfg. Co., 
Camden, New Jersey. 

The authors wish to express their thanks to the Reinhold 
Publishing Corporation for their care and consideration in mak- 
ing this second edition so attractive in spite of the abnormal con- 
ditions prevailing in the publishing field. 

A Bibliography of Electron Microscopy* 

Compiled by 

Claire Marton, Division of Electron Optics, Stanford University, 
Stanford University, California 


Samuel Sass, In Charge of Physics and Astronomy Libraries, University of Michigan, 

Ann Arbor, Michigan 


Bruche, E. and Scherzer, O. Geometrische Elektronenoptik. J Springer, Berlin, 1934. 
Marton, L. Le microscope electronique et ses applications; suivi de: Henriot, E., 

Uoptique electronique des systemes centres. Revue d'Optique, Paris, 1935. 
Starks, H. J. H. The electron microscope. Reports on progress in Physics, 2, 283-291. 

Physical Society of London, 1935. 
Bruche, E. and Henneberg, W. Geometrische Elektronenoptik. Ergebnisse der exakten 

Naturiuissenschaften 15, 365-421. J. Springer, Berlin, 1936. 
Busch, H. and Bruche, E. Beitrage zur Elektronenoptik. J. A. Barth, Leipzig, 1937. 
Klemperer, O. Electron Optics. Cambridge University Press, Cambridge, 1939. 
Myers, L. M. Electron Optics. Chapman and Hall, London, 1939. 
Ardenne, M. von. Elektronen Ubermikroskopie. J. Springer, Berlin, 1940. 
AEG Forschungs-Institut. Jahrbuch. Sonderheft: Ubermikroskop. J. Springer, Berlin, 

Klemperer, O. Electron microscopes of high magnification. Reports on progress in 

physics 7, 107-129. Physical Society of London, 1940. 
Borries, B. von and Ruska, E. Microscopy of high resolving power by means of fast 

electrons. Ergebnisse der exakten Naturiuissenschaften 19, 237-322. J. Springer, 

Berlin, 1941. 
Bruche, E. and Recknagel, A. Elektronengerate, Prinzipien und Systematik. J. 

Springer, Berlin, 1941. 
Anderson, T. F. The study of colloids with the electron microscope. In Kraemer, 

E. O., ed. Advances in colloid science, Vol. 1., pp. 353-390. Interscience, New 

York, 1942. 
Burton, E. F. and Kohl, W. H. The electron microscope. Reinhold, New York, 1942. 
Marton, L. The electron microscope in biology. Annual review of biochemistry, 

Vol. 12, pp. 587-614. Annual Reviews Inc., Stanford University. 
Ramsauer, C. (editor) , Das freie Elektron in Physik und Technik. J. Springer, Berlin, 

Ramsauer, C. Elektronenmikroskopie. Bericht iiber Arbeiten des A E-G-Forschungs- 

Instituts 1930 bis 1941. Second edition. J. Soringer, Berlin, 1942. 
Hawley, G. G. Seeing the invisible, Knopf, New York, 1945. 


Bruche, E. and Johannson, H. Cinematography of oxide cathodes using the electron 
microscope. Ann. d. Physik 15, 145-166 (1932). 

* Journal of Applied Physics, 14, 522-531 (1943) and 15, 575-579 (1944). 



Briiche, E. and Johannson, H. New investigations of cathodes with the electron 

microscope. Phys. Zeits. 33, 898-899 (1932). 
Knoll, M., Houtermanns, F. G., and Schulze, E. Emission distribution at glow cath- 
odes with the magnetic microscope. Zeits. f. Physik 78, 340-362 (1932). 
Briiche, E. and Johannson, H. Crystallographic investigations with the electron 

microscope. Zeits. f. techn. Physik 14, 487-488 (1933). 
Briiche, E. Electron microscopical reproduction by means of photoelectric electrons. 

Zeits. f. Physik 86, 448-450 (1933). 
Johannson, H. Electron optical immersion objective. Ann. d. Physik 18, 385-413 

Knoll, M. and Lubszinsky, G. Electron microscope pictures with secondary electrons. 

Phys. Zeits. 34, 671-674 (1933). 
Richter, E. F. Electron microscope observations of the migration of emitting material 

on oxide cathodes. Zeits. f. Physik 86, 697-709 (1933). 
Ruska, E. Formation of pictures of surfaces irradiated by electrons in the electron 

microscope. Zeits. f. Physik 83, 492-497 (1933). 
Benham, W. E. Demonstration of low voltage electron microscope using electrostatic 

focusing. /. E. E., J. 75, 388-390 (1934). 
Calbick, C. J. and Davisson, C. J. Electron microscope, Phys. Rev. 45, 764 (1934). 
Briiche, E. The electron microscope and its application in particular to the study of 

thin layers on metals. Kolloid Zeits. 69, 389-394 (1934). 
Briiche, E. and Knecht, W. Electron-optical observation of the transformation of iron 

at temperatures between 500° and 1000°C. Zeits. f. techn. Physik 15, 461-463 

Briiche, E. and Knecht, W. Note on attaining high resolution with the electron optical 

immersion objective. Zeits. f. Physik 92, 462-466 (1934). 
Johannson, H. Immersion objective for electron microscopes. Ann. d. Physik 21, 

274-284 (1934). 
Knecht, W. Combined light and electron microscope. Ann. d. Physik 20, 161-182 

(1934). . 
Briiche, E. The electron-optical image of structures and its interpretation in the 

emission ratio of barium-nickel cathodes. Zeits. f. Physik 98, 77-107 (1935). 
Briiche, E. and Knecht, W. Electron-optical observation of the a-y transformation in 

iron. Zeits. f. techn. Physik 16, 95-98 (1935). 
Briiche, E. and Mahl, H. On the emission image of thoriated tungsten and thoriated 

molybdenum. I. Zeits. f. techn. Physik 16, 623-627 (1935). 
Dosse, J. and Knoll, M. Electron distribution at focus of x-ray tubes with electron 

microscope. Arch. f. Elektrot. 29, 729-730 (1935). 
Kemnitz, G., Knoll, M., and Walcher, W. Electron concentration by concave hot 

cathode surfaces with the electron microscope. Zeits. f. Physik 96, 612-619 (1935). 
Mahl, H. Electron-optical image formation with emitting wires. Zeits. f. Physik 98, 

321-322 (1935). 
Schenk, D. Distribution of emission over a crystalline heated cathode. Ann. d. Physik 

23, 240-254 (1935). 
Behne, R. Characteristics of the immersion objective for image formation with fast 

electrons. Ann. d. Physik 26, 372-384 (1936). 
Behne, R. Contribution to the knowledge of the electron-optical immersion lens, II. 

Zeits. f. Physik 101, 521-526 (1936). 
Behne, R. Photography of foils with the electron-optical immersion objective. Ann. 

d. Physik 26,385-397 (1936). 
Briiche, E. and Mahl, H. On the emission image of thoriated tungsten and thoriated 

molybdenum, II. Zeits. f. techn. Physik 17, 81-84 (1936). 
Bruche, E. and Mahl, H. On the emission image of thoriated tungsten and thoriated 

molybdenum, III. Zeits. f. techn. Physik 17, 262-266 (1936). 
Heinze, W. and Wagener, S. The processes in the activation of oxide cathodes. Zeits. 

f. techn. Physik 17, 645-653 (1936). 
Mahl, H. Influence of oxygen on glow emission investigated by the electron micro- 
scope. Zeits. f. techn. Physik 17, 653-656 (1936). 


Burgers, W. G. Direct observation of structural changes at high temperature with 
the electron microscope. Zeits. f. Metallkunde 29, 250-251 (1937). 

Burgers, W. G. and van Amstel, J. J. A. P. Electron-optical observation of metal 
surfaces. Parts I and II. Physica 4, 5-22 (1937). 

McMillen, J. H. and Scott, G. H. A magnetic electron microscope of simple design. 
Rev. Sci. Instruments 8, 288-290 (1937). 

Burgers, W. G. and van Amstel, J. J. A. P. Electron-optical observation of the tran- 
sition of a- into |8- zirconium. Nature 141, 330 (1938). 

Mahl, H. Electron-optical examination of electronic and ionic emissions of wires. 
Zeits. f. Physik 108, 771-776 (1938). 

Meschter, E. An electron microscope for studying thermal and secondary electron 
emission. Rev. Sci. Instruments 9, 12-15 (1938). 

Scott, G. H. and Packer, D. M. The localization of minerals in animal tissues by the 
electron microscope. Science 89, 227-228 (1938). 

Scott, G. H. and Packer, D. M. The electron microscope as an analytical tool for 
the localization of minerals in biological tissues. Anat. Rec. 74, 17-30 (1939). 

Scott, G. H. and Packer, D. M. An electron microscope study of magnesium and 
calcium in striated muscle. Anat. Rec. 74, 31-40 (1939). 

Scott, G. H. An electron microscope study of calcium and magnesium in smooth 
muscle. Proc. Soc. Exp. Biol, and Med. 45, 30-31 (1940). 

Fox, G. W. and Bailey, F. M. Measurement of cathode emission by use of the elec- 
tron microscope. Phys. Rev. 59, 174-178 (1941). 

Boersch, H. Increase of resolution of the emission electron microscope. Naturwtss. 
30, 120 (1942). 

Boersch, H. Improvement of the resolving power of the emission electron micro- 
scope. Zeits. f. techn. Physik 23, 129-130 (1942). 

Briiche, E. The limit of resolution of the emission electron microscope. Kolloid 
Zeits. 100, 192-206 (1942). 

Dosse, J. and Muller, H. O. The resolving power of the emission electron microscope. 
Zeits. f. Physik 119, 415-422 (1942). 

Huber, H. and Wagener, S. X-ray and electron optical study of crystal structure of 
alkali earth oxide mixtures. Zeits. f. techn. Physik 23, 1-12 (1942). 

Mahl, H. An experiment in emission microscopy with electrostatic lenses. Zeits. f. 
techn. Physik 23, 117-119 (1942). 

Mecklenburg, W. Electrostatic emission electron microscope. Zeits. f. Physik 120, 
21-30 (1942). 

Recknagel, A. Resolving power of the emission electron microscope. Zeits. f. Physik 
120, 331-362 (1943). 


Knoll, M. and Ruska, E. The electron microscope, Zeits. f. Physik 78, 318-339 (1932) . 

Marton, L. and Nuyens, M. Geometrical electron optics. Wis-en Natuurkundig Tijd- 
schrift6, 159^170 (1933). 

Marton, L. Electron microscopy of biological objects. I. Bull. Acad. Roy. de Belg., 
CI. des Sciences 20, 439-446 (1934). 

Ruska, E. Advances in building and performance of the magnetic electron micro- 
scope. Zeits. f. Physik 87, 580-602 (1934). 

Ruska, E. Magnetic objective for the electron microscope. Zeits. f. Physik 89, 90-128 

Sommerfeld, A. and Scherzer, O. Description of an electron microscope. Munich. 

Med. Wochenschr. 81, 1859-1860 (1934). 
Knoll, M. The electron microscope. Zeits. f. Arztl. Fortbild. 32, 644-678 (1935). 
Marton, L. Electron microscopy of biological objects. III. Bull. Acad. Roy. de Belg., 

CI. des Sciences 21, 606-617 (1935). 
Fritz, R. The electron microscope. Rev. Gen. des Sciences 47, 338-342 (1936). 
Martin, L. C, Whelpton, R. V., and Parnum, D, H. A new electron microscope. 

Journ. Sci. Instruments 14, 14-24 (1937). 


Borries, B. von and Ruska, E. Development and present efficiency of the electron 

microscope. Wiss. Veroffent. Siemens-Werke 17, 99-106 (1938). 
Nuttall, A. K. An electron microscope for high magnifications. Metro-Vickers Gaz. 

17, 256-259 (1938). 
Ardenne, M. von. An electrostatic high voltage lens of short focal distance. Natur- 

wiss. 27, 614-615 (1939). 
Ardenne, M. von. Single crystal fluorescent screens for supermicroscopy. Zeits. f. 

techn. Physik 20, 235-239 (1939). 
Borries, B. von and Ruska, E. Development and efficiency of the Siemens ultramicro- 

scope. Z. f. whs. Mikroskop. 56, 317-333 (1939). 
Borries, B. von and Ruska, E. An electron supermicroscope for research institutes. 

Naturwiss. 27, 577-582 (1939). 
Briiche, E. and Haagen, E. A new simple supermicroscope and its application to 

bacteriology. Naturwiss. 27, 809-811 (1939). 
Burton, E. F., Hillier, J., and Prebus, A. A report on the development of the electron 

microscope at Toronto. Phys. Rev. 56, 1171-1172 (1939). 
Mahl, H. The electrostatic electron microscope. Zeits. f. techn. Physik 20, 316-317 

Martin, L. C. The electron microscope. /. Roy, Microscop. Soc. 59, 217-231 (1939). 
Martin, L. C, Parnum, D. H., and Speak, G. S. Report on experimental work on the 

development of the electron microscope. /. Roy. Microscop. Soc. 59, 203-216 

Prebus, A. and Hillier, J. Construction of a magnetic electron microscope of high 

resolving power. Canad. J. of Research 17, 49-63 (1939). 
Ardenne, M. von. Results of a new electron ultramicroscope. Naturwiss. 28, 113-127 

Ardenne, M. von. Stereo-ultramicroscopy with the universal electron microscope. 

Naturwiss. 28, 248-252 (1940) .. 
Ardenne, M. von. A universal electron microscope for bright field, dark field and 

stereoscopic use. Zeits. f. Physik 115, 339-368 (1940). 
Borries, B. von, and Ruska, E. The effect of the potential difference on the super- 
microscopic image. Zeits. f. Physik 116, 249-256 (1940). 
Kinder, E. and Pendzich, A. A new magnetic lens of short focal length. Jahrbuch 

AEG-Forsch., Sonderheft, Ubermikroskop 7, 23-26 (1940). 
Kinder, E. A new high-potential ultramicroscope. Zeits. f. techn. Physik 21, 222-223 

Mahl, H. The electrostatic electron ultramicroscope. Jahrbuch AEG-Forsch., Son- 
derheft, Ubermikroskop 7, 43-56 (1940). 
Martin, L. C. Ultraviolet and electron microscopy. Nature 146, 288-292 (1940). 
Marton, L. A new electron microscope. Phys. Rev. 58, 57-60 (1940). 
O'Daniel, H. and Radczewski, O. E. Electron microscopy and electron-diffraction 

study on the same specimen of highly dispersed minerals. Naturwiss. 28, 628-630 

Marton, L., Banca, M. C, and Bender, J. F. A new electron microscope. RCA Rev. 

5, 232-243 (1940). 
Prebus, A. Improved pole piece construction of the objective lens of a magnetic 

electron microscope. Canad. J. of Research 18, 175-177 (1940). 
Ruska, E. Electron diffraction patterns recorded in the supermicroscope. Wiss. 

Veroffentl. Siemens-Werke, Werkstoff -Sonderheft 372-379 (1940). 
Ruska, H. Supermicroscopy. Ned. T. Natuurk. 7, 179-191 (1940). 
Zworykin, V. K. An electron microscope for the research laboratory. Science 92, 

51-53 (1940). 
Ardenne, M. von. Electron ultramicroscopy with accessory probe for production of 

electron diffraction diagrams with very limited field of view. Zeits. f. Physik 117, 

515-523 (1941). 
Hillier, J. and Vance, A. W. Recent developments in the electron microscope. Proc. 

Inst. Radio Engrs. 29, 167-176 (1941). 
Hughes, A. L. The magnetic electron lens. Amer. J. Phys. 9, 204-207 (1941). 


Vance, A. W. Stable power supplies for electron microscopes. RCA Rev. 5, 

293-300 (1941). 
Zworykin, V. K., Hillier, J., and Vance, A. W. An electron microscope for practical 

laboratory service. Trans. Am. Inst. Elect. Engrs. 60, 157-162 (1941). 
Ardenne, M. von, Schiebold, E., and Gunther, F. Fine beam electron diffraction in 

the universal electron microscope. Zeits. f. Fhysik. 119, 352-365 (1942). 
Ardenne, M. von. Further development of the universal electron microscope. Phys. 

Zeits. 43, 11-15 (1942). 
Bachman, C. H. and Ramo, S. Simplified electron microscope. Phys. Rec. 62, 494 

Hillier, J., Baker, R. F., and Zworykin, V. K. A diffraction adapter for the electron 

microscope. /. of Applied Physics 13, 571-577 (1942). 
Bachman, C. H. and Ramo, S. Electrostatic electron microscopy. I. /. of Applied 

Physics 14, 8-18 (1943). 
Bachman, C. H. and Ramo, S. Electrostatic electron microscopy. II. /. of Applied 

Physics 14, 69-76 (1943). 
Bachman, C. H. and Ramo, S. Electrostatic electron microscopy. III. /. of Applied 

Physics 14, 155-160 (1943). 
Bachman, C. H. Simplified electron microscopy. Electronics 16, 78-81; 195-200 

Zworykin, V. K. Electron microscope: most recent research tool. Electronics 16, 

146-147; 254 (1943). 
Ardenne, M. von. The resolving power of photographic emulsions for electron 

beams. Zeits. f. Physik 114, 379 (1939). 
Borries, B. von and Ruska, E. Technique of the Siemens electron microscope. 

Siemens Zeits. 20, 217-227 (1940). 
Ardenne, M. von. Electron microscopy at elevated temperatures with the universal 

electron microscope. Kolloid Zeits. 97, 257-272 (1941). 
Briiche, E. The development of the electron microscope with electrostatic lenses. 

V.D.1. 85, 221-229 (1941). 
Lundgren, E. H. The electron microscope. Tek Tid. Uppl. A-c, Kemi 71, 20-35 

(1941). . . 

Boersch, H. Construction and possibilities of application of an electron microscopical 

arrangement. Phys. Zeits. 43, 515-520 (1942). 
Frey, F. Orientation marks on sample holders for the electron microscope. Zeits. f. 

techn. Physik 23, 82 (1942). 
Frey, F. Application and construction of a plateholder for two exposures with the 

Siemens electron microscope in its normal construction. Zeits. f. techn. Physik 

23, 176-177 (1942). 
Matthias, A. Remark on the origin of the electron microscope. Phys. Zeits. 43, 129- 

130 (1942). 
Prebus, A. The electron microscope. Ohio State Univ. Eng. Exp. Sta. News 14, 3 

Ruska, E. Taking of photographs with the electron microscope at high pressures. 

Kolloid Zeits. 100, 212-219 (1942). 
Ardenne, M. von. Electron microcinematography with the universal electron micro- 
scope. Zeits. f. Physik 120, 397-412 (1943). 
Briiche, E. The development of the electron microscope. Phys. Zeits. 44, 176-180 

Mahl, H. and Pendzich, A. Electron-mirror microscope: a new electron microscope. 

Zeits. f. techn. Physik 24, 38-42 (1943). 
Rtidenberg, R. The early history of the electron microscope. /. Appl. Phys. 14, 434- 

436 (1943). 
Zworykin, V. K. and Hillier, J. A compact high resolving power electron microscope. 
/. Appl. Phys. 14, 658-673 (1943). 



Davisson, C. J. and Calbick, C. J. Electron lenses. Phys. Rev. 38, 585 (1931). 
Davisson, C. J., and Calbick, C. J. Electron lenses. Phys. Rev. 42, 580 (1932). 
Knoll, M. and Ruska, E. Contribution to geometrical electron optics. I. Ann d 

Physik [5] 12, 607-640 (1932). 
Knoll, M. and Ruska, E. Contribution to geometrical electron optics. II. Ann. d. 

Physik [5] 12,461-661 (1932). 
Ruska, E. and Knoll, M. The magnetic concentration coil for fast electrons. I. 

Zeits. f. techn. Physik 12, 389^399 (1932). 
Ruska, E. and Knoll, M. The magnetic concentration coil for fast electrons. II. 

Zeits. f. techn. Physik 12, 448 (1932). 
Henneberg, W. The resolving power of the electron microscope for transparent 

objects. Zeits. f. Instrumentenk. 55, 300-305 (1935). 
Stabenov, G. Magnetic electron lens without image rotation. Zeits. f. Physik 96, 

612-619 (1935). 
Boersch, H. On the primary and secondary image in the electron microscope. I. 

Ann. d. Physik 26, 631-644 (1936). 
Boersch, H. On the primary and secondary image in the electron microscope. II. 

Ann. d. Physik 27, 75-80 (1936). 
Marton, L. Some considerations concerning the resolving power in electron micros- 
copy. Physica 3, 959-967 (1936). 
Ardenne, M. von. The image errors of the electron microscope caused by electron 

scattering in the object and their relation to each other. Zeits. f. Physik 11, 

152-157 (1938). 
Ardenne, M. von. Limits of the resolving power of the electron microscope. Zeits. 

f. Physik 108, 308-338 (1938). 
Borries, B. von, and Ruska, E. Image formation in the supermicroscope. Zeits. f 

techn*. Physik 19, 402-407 (1938). 
Ardenne, M. von. Chromatic errors in the electron microscope. Zeits. f. Physik 

113, 257-259 (1939). 
Ardenne, M. von. Intensity and resolving power of the electron microscope. Zeits. 

f. Physik 112, 744-753 (1939). 
Borries, B. von, and Ruska, E. Properties of the supermicroscopical image formation 

Naturwiss. 27, 281-287 (1939). 
Borries, B. von, and Ruska, E. The resolving power of the supermicroscope. Zeits 

f. techn. Physik 20, 225-235 (1939). 
Hillier, J. Effects of chromatic error on the electron microscope. Canad. J. of Re- 
search, A17, 64-69 (1939). 
Ardenne, M. von. Determination of the resolving power of electron microscope. 

Phys. Zeits. 42, 72-74 (1941) . 
Ardenne, M. von. Testing of short focus electron lens. Zeits. f. Physik 117, 602-611 

Hillier, J. A discussion of the fundamental limit of an electron microscope. Phys. 

Rev. 60, 743-745 (1941). 
Marton, L. and Schiff, L. I. Determination of object thickness in electron microscopy. 

/. of Applied Physics 12, 759-765 (1941). 
Ruska, E. Lenses for electron microscopes of high resolving power. Archiv. f. 

Elektrotechn. 36, 431-454 (1942). 
Schiff, L. I. Ultimate resolving power of the electron microscope. Phys. Rev. 61, 

721-722 (1942). 
Savchenko, F. The dependence of certain optical parameters on the electric and 

geometric parameters of electric immersion lens. Zhurnal Tekhnicheskoi Fiziki, 

Moscow, 9, 2211-2219 (1939). 
Borries, B. von and Ruska, E. The influence of electron interferences on the image 

formation of crystals in the electron microscope. Naturwiss. 28, 366-367 (1940). 
Dosse, J. Theoretical and experimental investigation of electron emitters. Zeits. f. 

Physik 115, 530-556 (1940). 


Ardenne, M. von. Supplement to the papers: "The testing of electron lenses of short 
focal length" and "a 200-kv universal electron microscope with an object screen." 
Zeits. f. Physik 118, 384-388 (1941). 

Glaser, W. and Lammel, E. For which electromagnetic fields is Newton's image equa- 
tion valid? Ann. d. Physik 40, 367-384 (1941). 

Glaser, W. On the distribution of the ampere-turn density of a given magnetic field 
produced by a cylindrical coil. Zeits. f. Physik 118, 264-268 (1941). 

Ardenne, M. von. The faithful representation of the structures close to the limit of 
resolution with the light microscope and the electron microscope. Kolloid Zeits. 
100, 206-211 (1942). 

Bertram, S. J. Calculation of axially symmetric fields. /. Appl. Phys. 13, 496-502 

Boersch, H. The influence of crystal lattice interferences on the imaging in the elec- 
tron microscope. Zeits. f. Physik 118, 706-713 (1942). 

Copeland, P. L. Thin electrostatic lenses for electrons. Am. J. Phys. 10, 236-246 

Craggs, J. D. The electrostatic focusing of high speed ion and electron beams. /. 
Appl. Phys. 13, 772-786 (1942). 

Glaser, W. Electron-optical imaging with distorted rotational symmetry. Zeits. f. 
Physik 120, 1-15 (1942). 

Boersch, H. Edge diffraction of electrons. Phys. Zeits. 44, 32-38 (1943). 

Boersch, H. Fresnel diffraction in the electron microscope. Phys. Zeits. 44, 202-211 

Glaser, W. Image formation and resolving power of the electron microscope from 
the point of view of wave mechanics. Zeits. f. Physik 121, 647-667 (1943). 

Wendt, G. The dioptrics of electron optical instruments with arbitrarily curved 
optical axis. Zeits. f. Physik 120, 720^741 (1943). 

Marton, L. and Hutter, R. G. E. The transmission type of electron microscope and its 
optics. Proc.l.R.E. 32, 3-12 (1944). 

Marton, L. and Hutter, R. G. E. On apertures of transmission type electron micro- 
scopes using magnetic lens. Phys. Rev. 65, 161-167 (1944). 

Opatowski, I. Refractive index in electron optics. Phys. Rev. 65, 54-55 (1944). 


Busch, H. Calculation of the trajectory of cathode rays in electromagnetic fields of 

axial symmetry. Ann, d. Physik 81, 974-993 (1926). 
Busch, H. On the action of the concentration coil in Braun tubes. Arch. f. Elektrot. 

18, 583-594 (1927). 
Picht, J. Theory of geometrical optics for electrons. Ann. d. Physik 15, 926-964 

Glaser, W. Geometrical optics of electron rays. Zeits. f. Physik 80, 451-464 (1933). 
Glaser, W. Geometrical electron optics of the axial symmetrical electromagnetic 

field. Zeits. f. Physik 81, 647-686 (1933). 
Glaser, W. Optical imagery by mechanical systems and the optics of nonhomogeneous 

isotropic media. Ann. d. Physik 18, 557-586 (1933). 
Glaser, W. Theory of the electron microscope. Zeits. f. Physik 83, 104-122 (1933). 
Johannson, H. and Scherzer, O. Electric electron condensing lens. Zeits. f. Physik 

80, 183-192 (1933). 
Posener, L. Theory of the electron microscope. Zeits. f. Physik 80, 813-818 (1933). 
Scherzer, O. Theory of electric electron condensing lenses. Zeits. f. Physik 80, 

193^202 (1933). 
Diels, K. and Knoll, M. Proof of the aberrations of an electron lens by point imagery. 

Zeits. f. techn. Physik 16, 621-623 (1935). 
Glaser, W. Theory of image defects of an electron microscope. Zeits. f. Physik 

97, 171-201 (1935). 
Wallauschek, R. and Bergmann, P. Theory of electron microscope using purely mag- 
netic fields. Zeits. f. Physik 94, 329-347 (1935). 


Scherzer, O. On some defects of electron lenses. Zeits. f. Physik 101, 593-603 (1936). 
Scherzer, O. The weak electric single lens of least spherical aberration. Zeits. f. 

Physik 101, 23-26 (1936). 
Diels, K. and Wendt, G. Eight third-order aberrations of magnetic electron lenses. 

Zeits. f. techn. Physik 18, 65-69 (1937). 
Rebsch, R. and Schneider, W. Spherical aberration of weak electron lenses. Zeits. 

f. Physik 107, 138-143 (1937). 
Rogowski, W. Defects in electron images. Arch. f. Elektrot. 31, 555-593 (1937). 
Rebsch, R. Theoretical resolving power of the electron microscope. Ann. d. Physik 

31, 551-560 (1938). 
Scherzer, O. The theoretically attainable resolving power of the electron microscope. 

Zeits. f. Physik 114, 427-434 (1939). 
Voit, H. Third-order electron-optical abberations. Zeits. f. Instrument enk. 59, 

71-82 (1939). 
Glaser, W. Chromatic aberration in electron lenses. Zeits. f. Physik 116, 56-57 

Glaser, W. On a magnetic field free from spherical aberration. Zeit. f. Physik 116, 

19-33 (1940). 
Rebsch, W. and Glaser, W. Aperture error of electron lenses. Zeits. f. Physik 116, 

729-735 (1940). 
Dosse, J. Optical constants of strong electron lenses. Zeits. f. Physik 117, 722-753 

Dosse, J. Precise computation of unsymmetrical field form H = H /1+ (z/a) 2 . 

Zeits. f. Physik 117, 316-321 (1941). 
Glaser, W. Precise computation of magnetic lenses with the field form H=H /1 + 

(z/a) 2 . Zeits. f. Physik 117, 285-315 (1941). 
Kompfner, R. On a method of correcting the spherical error of electron lenses, 

especially those employed with electron microscopes. Phil. Mag. 32, 410-416 

Recknagel, A. Theory of electrical electron microscope which is self-illuminating. 

Zeits. f. Physik 177, 689-708 (1941). 
Plass, G. N. Electrostatic electron lenses with a minimum of spherical aberration. 

/. of Applied Physics 13, 49-55 (1942). 
Ramberg, E. G. Variation of the axial aberrations of electron lenses with lens 

strength. /. of Applied Physics 13, 582-594 (1942). 
Ardenne, M. von. On the magnitude of the aperture defect in the electron micro- 
scope. Zeits. f. techn. Physik 20, 289-290 (1939). 
Dosse, J. Supplement to the paper: "Optical characteristics of strong electron lenses." 

Zeits. f. Physik 118, 375-383 (1941). 
Gobrecht, R. Experimental investigations of spherical aberration in electrostatic lenses. 

Arch. f. Elektrotechn. 35, 672-685 (1941). 
Scherzer, O. The lower limits of focal distance and of chromatic defect of magnetic 

electron lenses. Zeits. f. Physik 118, 461-466 (1941). 
Plass, G. N. Errata: Electrostatic electron lenses with a minimum of spherical aber- 
ration. /. Appl. Phys. 13, 524 (1942). 
Boersch, H. Secondary images in electron microscopy. Zeits. f. Physik 121, 746- 

754 (1943). 


Marton, L. The electron microscope in biology. Acta Union Intemat. Contre Cancer 
4,221-238 (1939). 

Ardenne, M. von. 200-kv universal electron microscope with an object screen. 
Zeits. f. Physik 117, 657-688 (1941). 

Miiller, H. O. and Ruska, E. A supermicroscope for 220-kv beam potential. Kolloid 
Zeits. 95, 21-25 (1941). 

Zworykin, V. K., Hillier, J., and Vance, A. W. A preliminary report on the develop- 
ment of a 300-kv magnetic electron microscope. /. of Applied Physics 12, 738- 
742 (1941). 


Ardenne, M. von. Supplement to the papers: "The testing of electron lenses of short 
focal length" and "A 200-kv universal electron microscope with an object screen." 
Zeits. f. Physik 118, 384-388 (1941). 


Johnson, R. P. and Shockley, W. Investigation of thermionic filaments with a simple 

electron microscope. Phys. Rev. 48, 973 (1935). 
Wehnelt A. and Schilling, W. Electron microscope examination of the electron 

emission of cold metals. Zeits. f. Physik 98, 286-287 (1935). 
Johnson, R. P. and Shockley, W. An electron microscope for filaments: emission and 

adsorption by tungsten single crystals. Phys. Rev. 49, 436-440 (1936). 
Miiller, E. W. Electron microscopical observations of field emittors. Zeit. f. Physik 

106, 541-550 (1937). 
Ardenne, M. von. The electron scanning microscope. Zeits. f. techn. Physik 19, 

407-416 (1938). 
Ardenne, M. von. Electron scanning microscope — theoretical fundamentals. Zeits. 

f. Physik 109, 553-573 (1938). 
Johnson, R. P. Simple electron microscopes. /. of Applied Physics 9, 508-516 

Boersch, H. The electron shadow microscope. I. Zeits. f. techn. Physik 20, 346- 

350 (1939). 
Zworykin, V. K., Hillier, J., and Snyder, R. L. A scanning electron microscope. A. S. 

T. M. Bull. 117, 15-23 (1942). 
Ruthemann, G. Discrete energy losses of fast electrons in solids. Naturwiss. 29, 648 

Haefer, R. Field electron microscopic observations on tungsten points. Zeits. Krist. 

104, 1-10 (1942). 
Ruthemann, G. Electron retardation at x-ray levels. Naturwiss. 30, 145 (1942). 
Hillier, J. On microanalysis by electrons. Phys. Rev. 64, 318-319 (1943). 
Miiller, E. W. Resolving power of the field electron microscope. Zeits. f. Physik 

120, 270-282 (1943). 
Miiller, E. W. Velocity distribution of electrons for field emission. Zeits. f. Physik 

120, 261-269 (1943). 


Borries, B. von, and Ruska, E. Images of films in transmission by means of the elec- 
tron microscope. Zeits. f. Physik 83, 187-193 (1933). 
Marton, L. The electron microscope; first attempts at its application to biology. 

Ann. et Bull, de la Soc. Roy. des Sciences Med. et Nat. de Bruxelles [5-6] 92, 

106 (1934). 
Marton, L. Electron microscopy of biological objects. Phys. Rev. 46, 527-528 

Marton, L. Electron microscopy of biological objects. Nature 133, 911 (1934). 
Borries, B. von, and Ruska, E. Electron microscope and its uses. V. D. I. 79, 519- 

524 (1935). 
Driest, E. and Miiller, H. O. Electron micrographs of chitin substances. Z. f. wiss. 

Mikrosk. 52, 53-55 (1935) . 
Marton, L. The electron microscope and its applications. Rev. d'Optique 14, 

129-145 (1935). 
Marton, L. The electron microscope and its applications especially in biology. 

Seme Congres Nat. d. Sciences, Bruxelles 2, 928-933, 1935. 
Marton, L. Electron microscopy of biological objects. II. Bull. Acad. Roy. de 

Belg., CI. des Sciences 21, 553-564 (1935). 
Saupe, E. Use of electron microscope and diffraction in biology and pathology. 

Frankfurt. Z. f. Path. 47, 485-516 (1935). 


Krause, F. Electron-optical pictures of diatoms with the magnetic electron micro- 
scope. Zeits. f. Physik 102, 417-422 (1936). 

Marton, L. The electron microscope. Rev. Microbiol. Appl. 3, 117-124 (1936). 

Marton, L. Electron microscopy of biological objects. IV. Bull. Acad. Roy. de 
Belg., CI. des Sciences 22, 1336-1344 (1936) . 

Beischer, D. and Krause, F. Electron microscope in colloid investigations. Naturiviss. 
25, 825-829 (1937). 

Krause, F. The electron microscope and its use in biology. Naturiviss. 25, 817-825 

Luft, F. Practical uses of electron optics. Rontgenpraxis 9, 384-394 (1937). 

Marton, L. Electron microscopy of biological objects. V. Bull. Acad. Roy. de 
Belg., CI. des Sciences 23, 672-678 (1937). 

Beischer, D. Determination of size of crystals in metal and metal oxide smokes by 
means of Rontgen- and electron-diffraction diagrams and electron photo- 
micrographs. Zeits. Elektrochem. 44, 375-385 (1938). 

Borries, B. von, Ruska, E., and Ruska, H. Bacteria and virus as seen in the ultra- 
microscope, with introduction to technology of ultramicroscopy. Klin. 
Wochenschr. 17, 921-925 (1938). 

Borries, B. von, Ruska, E., and Ruska, H. Photographs of bacteria with the super- 
microscope. Wiss. Veroffent. Siemens-Werke 17, 107-111 (1938). 

Borries, B. von, Ruska, E., and Ruska, H. Status of the supermicroscope. V. D. I. 
82, 937-941 (1938). 

Eitel, W. H., Miiller, H. O., and Radczewski, O. E. Examination of clay minerals 
with the ultramicroscope. Ber. Deut. Keram. Ges. 20, 165-180 (1938). 

Krause, F. Examination of viruses by the electron microscope. Naturiviss. 26, 122 

Marton, L. A note on the article: "Practical uses of electron optics," by Fritz Luft. 
Rontgenpraxis 10, 352-353 (1938). 

Beischer, D. and Krause, F. The electron microscope in colloid chemistry. Angeiv. 
Chem. 51, 331-334 (1938). 

Ruchardt, E. New developments in electron microscopy and their results. Munch. 
Med. Wochenschr. 85, 1832-1837 (1938). 

Ardenne, M. von. Possibility of examining living substances with electron micro- 
scopes. Zeits. f. techn. Physik 20, 239-242 (1939). 

Frank, F. and Ruska, H. Examination of the blue structure of bird feathers with the 
electron microscope. Naturiviss. 27, 229-230 (1939). 

Friess, H. and Miiller, H. O. Smokes and dusts in the electron microscope. Gasmaske 
11, 1-9 (1939). 

Kausche, G. A., Pfankuch, E., and Ruska, H. Examination of plant viruses with the 
high power electron microscope. Naturiviss. 27, 292-299 (1939). 

Kausche, G. A. and Ruska, H. The structure of the "crystalline aggregates" of tobacco 
mosaic virus proteins. Biochem. Z. 303, 221-230 (1939). 

Kausche, G. A. and Ruska, H. Supermicroscopic investigation of the adsorption of 
the metal colloids on proteins. Kolloid Zeits. 89, 21-26 (1939). 

Krause, F. Use of the electron microscope in biology and medicine. Arch. Exp. 
Zellforsch. 22, 668-672 (1939). 

Marton, L. On the sensitivity of photographic emulsions for electrons between 50 
and 100 kv. Phys. Rev. 56, 290 (1939). 

Miiller, H. O. The supermicroscope in medicine. Med. Klinik 35, 1041-1043 (1939). 

Piekarski, G. and Ruska, H. Supermicroscopic pictures of bacterial flagellae. Klin. 
Wochenschr. 18, 383-386 (1939). 

Piekarski, G. and Ruska, H. Supermicroscopical investigations of bacteria with 
particular reference to the so-called nucleoids. Arch. Mikrobiol. 10, 302-321 

Radczewski, O. E., Miiller, H. O., and Eitel, W. The hydration of tricalcium 
aluminate. Naturiviss. 27, 837-838 (1939). 

Radczewski, O. E., Miiller, H. O., and Eitel, W. Supermicroscopical investigation of 
the hydration of lime. Zement 28, 693-697 (1939). 


Radczewski, O. E., Miiller, H. O., and Eitel, W. The hydration of tricalcium silicate. 

Naturwiss. 27, 807 (1939). 
Ruska, H. Supermicroscopic demonstration of organic structure. Arch. Exp. Zell- 

forsch. 22, 673-680 (1939). 
Ruska, H., Borries, B. von, and Ruska, E. The importance of the supermicroscope in 

virus research. Arch. Ges. Virusforsch. 1, 155 (1939). 
Wolpers, C. and Ruska, H. Structural investigations of the clotting of blood. Klin. 

Wochenschr. 18, 1077-1111 (1939). 
Ardenne, M. von. Analysis of the structure of highly and very highly illuminated 

silver bromide grains with the universal electron microscope. Z. Angew. Photogr. 

2, 14-20 (1940). 
Ardenne, M. von, and Beischer, D. Examination of the fine structure of high- 
molecular compounds with the universal electron microscope. I. The structure 

of /3-polyoxymethylene crystals. Zeits. f. ploys. Chem., [B] 45, 465-473 (1940). 
Ardenne, M. von, and Beischer, D. Examination of the fine structure of high- 
molecular compounds with the universal electron microscope. II. The morph- 
ology of rubber and Buna. Kautschuk 16, 55-60 (1940). 
Ardenne, M. von, Endell, K., and Hoffmann, E. Examination of finest fractions of 

bentonite and cement with the universal electron microscope. Ber. Deut. Keram. 

Ges. 21, 209-227 (1940). 
Ardenne, M. von. Images of minute particles, especially of molecules, by means of 

the universal electron microscope. Zeits. f. phys. Chem. [A] 187, 1-12 (1940). 
Ardenne, M. von, and Beischer, D. Investigation of catalysts with the universal 

electron microscope. Angew. Chem. 53, 103-107 (1940). 
Ardenne, M. von, and Beischer, D. Investigation of metal oxide smokes with the 

universal electron microscope. Zeits. Elektrochem. 46, 270-277 (1940). 
Ardenne, M. von. Occurrence of black lines in electron microscope images of crystal 

lamellae. Zeits. f. Physik 116, 736-738 (1940). 
Ardenne, M. von. Stage vibrator for electron microscopes. Kolloid Zeits. 93, 158- 

163 (1940). 
Borries, B. von, and Kausche, G. A. Determination of the size and shape distribution 

of colloidal gold by means of the supermicroscope. Kolloid Zeits. 90, 132-141 

Borries, B. von, and Ruttmann, W. Metallographic investigations of steel, cast iron, 

and brass by means of the supermicroscope. Wiss. Veroffentl. Siemens-Werke, 

Werkstoff-Sonderheft 342-362 (1940). 
Borries, B. von. Higher resolving power at the formation of images of surfaces by the 

supermicroscope. Zeits. f. Physik 116, 370-378 (1940). 
Burton, E. F., Hillier, J., and Prebus, A. Contribution of the electron microscope to 

medicine. Canad. M. A. Jour. 42, 116-119 (1940). 
Columbian Carbon Company Research Laboratories. The particle size and shape of 

colloidal carbon as revealed by the electron microscope. Columbian Colloidal 

Carbons 2, 5-53 (1940). 
Eitel, W. and Schusterius, C. The determination of the efficient surfaces of clay 

particles with the supermicroscope. Chem. Erde 13, 322-335 (1940). 
Eitel, W. and Schusterius, C. The interpretation of electron microscope pictures for 

the determination of clay grain distribution. Naturwiss. 28, 300-303 (1940). 
Eitel, W. and Radczewski, O. E. The characteristics of the clay mineral montmoril- 

lonite as shown by the supermicroscope. Naturwiss. 28, 397-399 (1940). 
Eitel, W. and Gotthardt, E. Stereophotographic measurement of the thickness of 

very small crystals by electron microscope observation. Naturwiss. 28, 367 

Friihbrodt, E. and Ruska, H. Research on the structure of bacteria, especially of 

bacterium membranes and capsules. Arch. Mikrobiol. 11, 137-154 (1940). 
Husemann, E. and Ruska, H. Making glycogen molecules visible. /. Prakt. Chem. 

156, 1-10 (1940). 
Husemann, E. and Ruska, H. Making visible the molecules of p-iodobenzoyl-glyco- 

gen. Naturwiss. 28,534 (1940). 


Jakob, A. and Mahl, H. Application of the supermicroscope in bacteriology, espe- 
cially for experiments with capsules. Jahrbuch AEG-Forsch., Sonderheft, Uber- 

mikroskop 7, 77-87 (1940). 
Jakob, A. and Mahl, H. Structure of bacteria, in particular capsulation of anarobiae 

by means of the electrostatic supermicroscope. Arch. Exp. Zellforsch. 24, 87-104 

Kausche, G. A. and Ruska, H. The identification of tobacco mosaic virus molecules 

in the chloroplast of virus-infected plants. Naturiviss. 28, 303 (1940). 
Kausche, G. A. Investigations on the problem of biological characterization of phyto- 

pathogenic virus proteins. Arch. Ges. Virus] 'orsch. 1, 362-372 (1940). 
Kausche, G. A. The mechanism of the gold-sol reaction of the protein of tobacco 

mosaic and potato-X-virus. Biol. Zentr. 60, 179-199 (1940). 
Kausche, G. A. and Ruska, H. The problem of the structure of the chloroplast. 

Naturiviss. 28, 303-304 (1940). 
Kausche, G. A. Results and problems of experimental plant virus research. Ber. Deut. 

Botan. Ges. 58, 200-222 (1940). 
Koch, L. and Lehmann, A. Supermicroscopical investigation of rolled aluminum 

surfaces. Wiss. Ver offend. Siemens-Werke, Werkstoff -Sonderheft, 363-371 

Lembke, A., Ruska, H., and Christophersen, J. Comparative microscopical and 

supermicroscopical observations on the infective agent of tuberculosis. Klin. 

Wochenschr. 19, 217-220 (1940). 
Mahl, H. Application of the supermicroscope in colloid chemistry and metallurgy. 

Jahrbuch AEG-Forsch., Sonderheft. Ubermikroskop 7, 67-76 (1940). 
Mahl, H. Determination of the orientation of single crystals of aluminum by means 

of the supermicroscope. Metallwirtschaft 19, 1082-1085 (1940). 
Mahl, H. The electrostatic electron microscope and some applications in colloid 

chemistry. Kolloid Zeits. 91, 105-117 (1940). 
Mahl, H. Metallurgical researches with the electrostatic ultramicroscope. Zeits. f. 

techn. Physik 21, 17-18 (1940 ! ). 
Mahl, H. A plaster-cast method for supermicroscopical investigation of metal sur- 
faces. Metallwirtschaft 19, 488-491 (1940). 
Meldau, R. Fine dusts in the submicroscopic field. V. D. I. 84, 103-106 (1940). 
Middel, V., Reichmann, R., and Kausche, G. A. Supermicroscopic investigation of 

the structure of bentonites. Wiss. Veroffentl. Siemens-Werke, Werkstoff- 

Sonderheft 334-341 (1940). 
Pfankuch, E. and Kausche, G. A. Ultramicroscopic examination of bacteriophages. 

Naturwiss. 28, 46 (1940). 
Radczewski, O. E., Midler, H. O., and Eitel, W. Supermicroscopic investigation of 

the beginning of precipitation of calcium carbonate from aqueous solutions. 

Zentr. Mineral. Geol. 8-19 (1940). 
Ruska, H. Making visible the bacteriophage lysis in the supermicroscope. Naturwiss. 

28, 45-46 (1940). 
Ruska, E. and Miiller, H. O. Progress in reproduction of electron irradiated surfaces. 

Zeits. f. Physik 116, 366-369 (1940). 
Ruska, H. The structure of cellulose fibers. Kolloid Zeits. 92, 276-285 (1940). 
Ruska, H. and Wolpers, C. The structure of "liquor-fibrin." Klin. Wochenschr. 19, 

695 (1940). 
Ruska, H. and Kretschmer, M. Supermicroscopic investigation of the degradation of 

cellulose fibers. Kolloid Zeits. 92, 163-166 (1940). 
Ruska, H. and Friihbrodt, E. Supermicroscopy as a method of investigation. Biologe 

9, 69^75 (1940). 
Ruska, H. Visibility of bacteriophage cultures in the ultramicroscope. Naturwiss. 

28, 45-46 (1940). 
Trurnit, H. and Friedrich-Freksa, H. The electron-microscopical investigation of 

"tomato mosaic virus Dahlem 1940." Biol. Zentr. 60, 546-556 (1940). 
Zahn, H. Experiments on the supermicroscopy of wool. Melliand Textilberichte 

21, 505-508 (1940). 


Anderson, T. F. and Stanley, W. M. A study of the reaction between the tobacco 

mosaic virus and its antiserum by means of the electron microscope. /. Biol. 

Chem. 139, 339-344 (1941). 
Ardenne, M. von. Electron microscopy of living tissue. Naturwiss. 29, 521-523 

Ardenne, M. von, and Friedrich-Freksa, H. Germination of the spores of the bacillus 

vulgatus from photographs in the 200-kv universal electron microscope. Natur- 
wiss. 29, 523-528 (1941). 
Barnes, R. B. and Burton, C. J. Metallic smokes as test objects in electron microscopy. 

Ind. Eng. Chem., News Ed. 19, 965-967 (1941). 
Beischer, D. Newer methods for investigating the structure of colloidal systems. 

Kolloid. Zeits. 96, 127-135 (1941). 
Borries, B. von, and Janzen, S. Image of finely machined technical surfaces in the 

supermicroscope. V. D. I. 85, 207-211 (1941). 
Hall, C. E. and Schoen, A. L. Application of the electron microscope to the study 

of photographic phenomena. /. Opt. Soc. Am. 31, 281-285 (1941). 
Hass, G. and Kehler, H. Investigations on the electrolytically produced and annealed 

layers of aluminum oxide by means of electron interference and measurements 

in the supermicroscope. Kolloid Zeits. 97, 27-35 (1941). 
Hass, G. and Kehler, H. On a temperature resistant durable carrier for electron 

interference and electron-microscope investigations. Kolloid Zeits. 95, 26-29 

Humbert, R. P. and Shaw, B. Studies of clay particles with the electron microscope. 

Soil Science 52, 481-487 (1941). 
Knoll, M. Evidence for built-up layers of iron oxide with the electron microscope. 

Phys. Zeits. 42, 120-122 (1941). 
Kregel, E. A. et al. A technique for electron-microscopic examination of encapsulated 

bacteria. Science 94, 592 (1941). 
Marton, L. The electron microscope. /. Bact. 41, 397-413 (1941). 
Marton, L., McBain, J. W., and Void, R. D. Electron microscope study of curd fibres 

of sodium laurate. /. Am. Chem. Soc. 63, 1990-1993 (1941). 
Mehl, R. F. The structure and rate of formation of pearlite. Trans. Am. Soc. Metals 

29, 813-862 (1941). 
Menke, W. Investigations of the fine structure of protoplasm with the universal 

electron microscope. Protoplasma 35, 115 (1941). 
Morton, H. E. and Anderson, T. F. Electron-microscopic studies of biological reac- 
tions. I. Reduction of potassium tellurite by Coryne-bacterium diphtheriae. 

Proc. Soc. Exp. Biol. Med. 46, 272-276 (1941). 
Mudd, S. and Anderson, T. F. Demonstration by the electron microscope of the 

combination of antibodies with flagellar and somatic antigens. Journ. of Im- 
munology 42, 251-266 (1941). 
Mudd, S., Polevitzky, K., Anderson, T. F., and Chambers, L. A. Bacterial morphology 

as shown by the electron microscope. /. Bact. 42, 251-264 (1941). 
Mudd, S. and Lackman, D. B. Bacterial morphology as shown by the electron micro- 
scope. J. Bact. 41, 415-420 (1941). 
Mudd, S. The electron microscope. Frank. Inst. J. 231, 496-498 (1941). 
Midler, F. H. Fiber production, fiber characteristics, and molecular structure. 

Phys. Zeits. 42, 123-129 (1941). 
Schmieder, F. Electron microscopic investigation of the relation between covering 

quality and grain size of pigments. Kolloid Zeits. 95, 29-33 (1941). 
Shaw, B. T. and Humbert, R. P. Electron micrographs of clay minerals. Soil. Sci. 

Soc. Am. Proc. 6, 146-149 (1941). 
Stanley, W. M. and Anderson, T. F. A study of purified viruses with the electron 

microscope. /. Biol. Chem. 139, 325-338 (1941). 
Wallner, F. E. and Schiebold, E. Contribution to the interpretation of supermicro- 

scopical pictures of fibrous specimens. Kolloid Zeits. 97, 36-37 (1941). 
Wesch, L. Electron-optical properties of Lenard phosphors. Ann. d. Physik 40, 

249-294 (1941). 


Zworykin, V. K. and Ramberg, E. G. Surface studies with the electron microscope. 
/. of Applied Physics 12, 692-695 (1941). 

Anderson, T. F. and Richards, A. G. Electron microscope studies of some structural 
colors of insects. /. of Applied Physics 13, 748-758 (1942). 

Ardenne, M. von. Reaction-chamber supermicroscopy by means of the universal 
electron microscope. Zeits. f. phys. Chemie, [B] 52, 61-71 (1942). 

Baker, R. F., Ramberg, E. G., and Hillier, J. The photographic action of electrons in 
the range between 40 and 212 kilovolts. /. of Applied Physics 12, 450-456 

Barnes, R. B. and Burton, C. J. The electron microscope and its uses. A. S. T. M 
Bull. 116, 34-41 (1942). 

Brown, O. J., Jr. and Smith, W. R. Relation of paint properties to surface areas of 
carbon black. Ind. and Eng. Chem. Ind. Ed. 34, 352-355 (1942). 

Burton, C. J., Barnes, R. B., and Rochow, T. G. The electron microscope: calibra- 
tion and use at low magnifications, hid. and Eng. Chem. Ind. Ed. 34, 1429-1436 

Clark, G. L., Bernays, P. M., and Tordella, J. P. Progress in lime investigation with 
x-rays and electron microscope. Nat. Lime Assoc. Proc. 24th Annual Convention, 
May, 1942. 

Clark, G. L., Barnes, M. R., and Baylor, E. R. A study of lampbrush chromosomes 
by the electron microscope. Science 95, 250 (1942). 

Eisenhut, O. and Kuhn, E. Light microscopic and electron microscopic investiga- 
tions of native and artificial cellulose fibers. Angew. Chem. 55, 198-206 (1942). 

Frampton, L. V. The size and shape of the tobacco mosaic virus protein particle. 
Science 95, 232-233 (1942). 

Green, R. H., Anderson, T. F., and Smadel, J. E. Morphological structure of the virus 
of vaccinia. /. Exp. Med. 75, 651-656 (1942). 

Hall, C. E., Jakus, M. A., and Schmitt, F. O. Electron microscope observations of 
collagen. /. Am. Chem. Soc. 64, 1234 (1942). 

Hamly, D. H. and Watson, J. H. L. Electron and optical microscope interpretation 
of the wall of pleurosigma angulatnm. J. Opt. Soc. Am. 32, 433-442 (1942). 

Heidenreich, R. D. Electron reflections in MgO crystals with the electron micro- 
scope. Phys. Rev. 62, 291-292 (1942). 

Hillier, J. and Baker, R. F. The observation of crystalline reflections in electron 
microscope images. Phys. Rev. 61, 722-723 (1942). 

Luria, S. E. and Anderson, T. F. The identification and characterization of bac- 
teriophages with the electron microscope. Proc. Natl. Acad. Sci. 28, 127-130 

Marshall, C. E., Humbert, R. P., Shaw, B. T., and Caldwell, O. G. Studies of clay 
particles with the electron microscope. II. The fractionation of beidellite, non- 
tronite, magnesium bentonite, and attapulgite. Soil Science 54, 149-157 (1942). 

Marton, L. Applications of the electron microscope in colloid chemistry. Jour. 
Phys. Chem. 46, 1023-1032 (1942). 

Marton, L. and McBain, J. W. The electron microscope and its developments. 
Chemical Products 5, 61-67 (1942). 

Morton, H. E. and Anderson, T. F. Some morphologic features of the Nichols strain 
or treponema pallidum as revealed by the electron microscope. Am. J. Syph. 
Gono. and Ven. Dis. 26, 565-573 (1942). 

Mudd, S., Polevitzky, K., Anderson, T. F., and Kast, C. Bacterial morphology as 
shown by the electron microscope. III. Cell wall and protoplasm in a strain 
of fusobacterium. /. Bact. 44, 361-366 (1942). 

Mudd, S., Polevitzky, K., and Anderson, T. F. Bacterial morphology as shown by the 
electron microscope. IV. Structural differentiation within the bacterial proto- 
plasm. Arch. Path. 34, 199-207 (1942). 

Mudd, S. and Anderson, T. F. Selective staining for electron microscopy: the effects 
of heavy metal salts on individual bacterial cell. /. Exp. Med. 76, 103-108 (1942). 

Mueller, H. and Sakmann, B. W. Electro-optical properties of colloids. /. Opt. Soc. 
Am. 32, 309-317 (1942). 


Picard, R. G. Studies on the structure of thin metallic films by means of the electron 
microscope. Dissertation, U. of Michigan, 1942. (Manuscript) 

Richards, A. G., Jr. and Anderson, T. F. Electron micrographs of insect trachaea. 
/. N. Y. Ento. Soc. 50, 147-167 (1942). 

Richards, A. G., Jr. and Anderson, T. F. Electron microscope studies of insect cuticle 
with a discussion of the application of electron optics to this problem. /. Morph. 
71, 135-183 (1942). 

Richards, A. G., Jr. and Anderson, T. F. Further electron microscope studies on 
arthropod trachaea. /. N. Y. Ento. Soc. 50, 245-247 (1942). 

Richards, A. G., Jr., Anderson, T. F., and Hance, R. T. A microtome sectioning tech- 
nique for electron microscopy illustrated with sections of striated muscle. Proc. 
Soc. Exp. Biol, and Med. 51, 148-152 (1942). 

Rosenblatt, M. B., Fullam, E. F., and Gessler, A. E. Studies of mycobacteria with 
the electron microscope. Amer. Rev. of Tuberculosis 46, 587-598 (1942). 

Schaefer, V. J. New methods for preparing surface replicas for microscopic observa- 
tion. Floys. Rev. 62, 495-496 (1942). 

Schaefer, V. J. and Harker, D. Surface replicas for use in the electron microscope. 
/. of Applied Physics 13, 427-433 (1942). 

Schmitt, F. O., Hall, C. E., and Jakus, M. A. Electron microscope investigations of 
the structure of collagen. /. Cell, and Comp. Physiol. 20, 11-53 (1942). 

Scott, G. H. and Anderson, T. F. A study of connective tissue with the electron 
microscope. Anat. Rec. 82, 445 (1942). 

Sears, G. R. and Kregel, E. A. Application of the electron microscope to the prob- 
lems of the pulp, paper and paperboard industry. Paper Trade Jour. 114, 43-49 

Sharp, D. G., Taylor, A. R., Beard, D., and Beard, J. W. Electron micrography of 
the western strain equine encephalomyelitis virus. Proc. Soc. Exp. Biol, and Med. 
51, 206-207 (1942). 

Sharp, D. G., Taylor, A. R., Beard, D., and Beard, J. W. Study of the papilloma 
virus protein with the electron microscope. Proc. Soc. Exp. Biol, and Med. 50, 
205-207 (1942). 

Shaw, B. T. The nature of colloidal clay as revealed by the electron microscope. 
Jour. Phys. Chem. 46, 1032-1043 (1942). 

Stanley, W. M. and Anderson, T. F. Electron micrographs of protein molecules. 
/. Biol. Chem. 146, 25-30 (1942). 

Taylor, A. R., Sharp, D. G., Beard, D., and Beard, J. W. Electron micrography of 
the eastern strain equine encephalomyelitis virus. Proc. Soc. Exp. Biol, and Med. 
51, 332-334 (1942). 

Umbreit, W. W. and Anderson, T. F. A study of thiobacillus thiooxidans with the 
electron microscope. /. Bact. 44, 317-320 (1942). 

Wergin, W. What does electron microscopy reveal regarding the structure of cellu- 
lose fibers? Kolloid Zeits. 98, 131-141 (1942). 

Wiegand, W. B. and Ladd, W. A. Colloidal carbon as revealed by the electron 
microscope. Rubber Age 50, 431-436 (1942). 

Wile, U. J., Picard, R. G., and Kearny, E. B. The morphology of spirochaeta 
pallida in the electron microscope. /. Am. Med. Assoc. 119, 880-881 (1942). 

Alexander, L. T., Faust, G. T., Hendricks, S. B., Isley, H., and McMurdie, H. F. 
Relationships of the clay minerals halloysite and endellite. Am. Miner. 28, 1-18 

Barnes, R. B. and Burton, C. J. The electron microscope and cellulose. Ind. and Eng. 
Chem. Ind. Ed. 35, 120-125 (1943). 

Chambers, L. A. and Henle, W. Studies on the nature of the virus of influenza: 
I. The dispersion of the virus of influenza A in tissue emulsions and in extra- 
embryonic fluids of the chick. /. Exp. Med. 77, 251-264 (1943). 

Chambers, L. A., Henle, W., Lauffer, M. A., and Anderson, T. F. Studies on the 
nature of the virus of influenza: II. The size of the infectious unit in influenza A. 
/. Exp. Med. 77, 265-276 (1943). 


Heidenreich, R. D. and Peck, V. G. Fine structure of metallic surfaces with the 
electron microscope. /. of Applied Physics 14, 23-29 (1943). 

Mahl, H. and Stranski, J. N. The behavior of aluminium in metallographic etching. 
Naturwiss, 31, 12-17 (1943). 

Plotz, H., Smadel, J. E., Anderson, T. F., and Chambers, L. A. Morphological struc- 
ture of rickettsiae. J. Exp. Med. 77, 355-358 (1943). 

Richards, A. G., Jr., Steinbach, H. B., and Anderson, T. F. Electron microscope 
studies of squid giant nerve axoplasm. /. Cell, and Comp. Phys. 21, 129-137 

Zworykin, V. K. Electron microscope in chemistry. Electronics 16, 190-196 (Jan- 
uary, 1943) . 

Ardenne, M. von. The wedge sectioning method for producing microtome cuts of 
less than 10~ 3 mm thickness for purposes of electron microscopy. Zeits. f. wis- 
sensch. Mikr. 56, 8-23 (1939). 

Ardenne, M. von. Experiments in photographing the virus of the hoof and mouth 
disease by means of the universal electron microscope. Naturwiss. 28, 531-532 

Mahl, H. Electron microscope images of objects by transmitted radiation and of 
metal surfaces. Zeits. angew. Photogr. 2, 58-63 (1940). 

Radczewski, O. E., Miiller, H. O., and Eitel, W. Electron microscopic investigations 
of the hydration of lime. Veroffentl. Kaiser Wilhelm-lnst. Silikatforsch. 10, 
139-142 (1940). 

Ruska, H. Significance and results of the electron microscope. Siemens Zeits. 20, 
228-234 (1940). 

Ardenne, M. von. Experiments on making visible molecular roughness of crystal 
edges inclined to the lattice planes in the universal electron microscope. Naturwiss. 
29, 780-781 (1941). 

Ardenne, M. von and Weber, H. H. Electron microscopic investigations of the muscle 
protein "myosin." Kolloid Zeits. 97, 322-325 (1941). 

Endell, J. Significance of the electron microscope for identifying the microstructure 
of clays. Tonind. Ztg. 65, 69-72 (1941). 

Henneberg, W. Electron microscope, supermicroscope and metallography. Stahl u. 
Eisen 41, 769-777 (1941). 

Koch, L. and Lehmann, A. Electron microscope investigation of smooth aluminum 
surfaces. Aluminium 23, 304-309 (1941). 

Mahl, H. The electrostatic electron microscope and some results from the metal- 
lurgical field. Zeits. f. Metallk. 33, 68-73 (1941). 

Mahl, H. Electron microscopic detection of metallic precipitation by the replica 
method. Metallwirtschaft 20, 983-986 (1941). 

Mahl, H. The damaging of cellulose fibers by radiation in the electron microscope. 
Kolloid Zeits. 96, 7-10 (1941). 

Meldau, R. and Teichmuller, M. The morphology of finest lead-oxide sublimates. 
III. Zeits. f. Elektrochem. 47, 630-634 (1941). 

Meldau, R. and Teichmuller, M. The morphology of finest lead-oxide sublimates. IV. 
Zeits. f. Elektrochem. 47, 634-636 (1941). 

Ruska, H. Technique and research results of electron microscopy. Vierteljrsch. d. 
Naturf. Ges. Zurich 86, 20-21 (1941). 

Ardenne, M. von. Recent research on the universal electron microscope. Forschng. 
u. Fortschr. 18, 32-35 (1942). 

Ardenne, M. von. On an electron microscopic investigation of the structure of 
reflection reducing layers and on the determination of the dimensions of such 
layers. Zeits. angew. Photogr. 4, 15-16 (1942). 

Borries, B. von. On the intensity relations in the electron microscope. I. The black- 
ening of photographic plates by electron beams. Phys. Zeits. 43, 190-204 (1942). 

Borries, B. von. On the intensity relations in the electron microscope. II. Magnifi- 
cation, graininess, and resolving power of electron exposed photographic plates. 
Zeits. angew. Photogr. 4, 3-4 (1942). 


Borries, B. von. On the intensity relations in the electron microscope. III. Suitability 

and limits of sensitivity of photographic plates for electron microscope pictures. 

Zeits. f. Physik 119, 498-521 (1942). 
Duffek, V. and Mahl, H. The electron microscope representation of metallic surfaces 

by the replica process without injury to the surface of the sample. Arch. Eisen- 

huttenw. 16, 73-76 (1942). 
Fischer, H. and Kurz, F. Electron microscope images of anodic oxide films on 

aluminum and their growth. Korros. und Metallschutz 18, 42-50 (1942). 
Gotthardt, E. Stereoscopic measurements of objects with the electron microscope. 

Zeits. f. Physik 118, 714-717 (1942). 
Gundermann, J. and Kiilz, H. Observations on zinc black by means of the electron 

microscope. Kolloid Zeits. 98, 287-289 (1942) . 
Haam, E. von and Schuh, M. Cytologic studies with the electron microscope. /. 

Tech. Methods 22, 75-80- (1942). 
Heidenreich, R. D. and Peck, V. G. Electron microscope study of surface structure. 

Phys. Rev. 62, 291-293 (1942). 
Mahl, H. The electron microscopic representation of surfaces with the replica 

method. Naturwiss. 30, 207-217 (1942). 
Mahl, H. and Pawlek, F. Electron microscopic structural investigations of non- 
alloyed steels. Arch. Eisenhiittenw. 16, 219-222 (1942}.. 
Mahl, H. and Pawlek, F. Electron microscopic investigation of Al alloys. Zeits. f. 

Metallk. 34, 232-236 (1942). 
Miiller, H. O. Measurement of the depth of electron microscopic objects. Kolloid 

Zeits. 99, 6-28 (1942). 
Miiller, H. O. and Pasewaldt, C. W. Observations on the fine structure of the test 

diatom Pleurosigma angulatum W.Sm. and stereoscopic pictures with the electron 

microscope. Naturwiss. 30, 55-60 (1942). 
Rawlins, T. E. Recent evidence regarding the nature of viruses. Science 96, 425- 

426 (1942). 
Roberts, E. A. The structure of chloroplasts, chromoplasts, leucoplasts, and carotin 

"crystals" and their relationship to cellulose "particles" and colloidal carbon. 

Am. J. of Botany 29, Suppl. p. 16 (1942). 
Riidiger, O. Use of beryllium for electron microscopic replica foil. Naturwiss. 30, 

279 (1942). 
Semmler-Alter, E. Electron microscope investigation of martensite in different 

tempering stages in steel with 0.24 percent carbon. Arch. Eisenhiittenw. 16, 

223-225 (1942). 
Smakula, A. Remark on the paper of M. von Ardenne, "On an electron-microscopic 

investigation of the structure of reflection reducing layers and the determination 

of such layers." Zeits. angew. Photogr. 4, 15-16 (1942). 
Tiselius, A. and Gard, S. Electron microscopic observations on poliomyelitis virus 

preparations. Naturwiss. 30, 728-731 (1942). 
Winterfeld, E. von. The universal electron microscope. Recent results of electron 

microscopy. Messteclm. 18, 98-102 (1942). 
Ardenne, M. von. Suitability of gelatine-free photographic film for electron micros- 
copy. Zeits. f. Physik 121, 1-6 (1943). 
Ardenne, M. von and Beischer, D. Electron microscope photographs of fine reaction 

layers on crystals of metal oxide smokes. Kolloid Zeits. 102, 127-131 (1943). 
Baker, R. F., Ramberg, E. G., and Hillier, J. Erratum: The photographic action of 

electrons in the range between 40 and 212 kilovolts. /. Appl. Phys. 14, 39 (1943). 
Barrett, C. S. Metallography with the electron microscope. Metals Technology 10, 

1-19 (1943). 
Bole, G. A. and Loomis, G. A. Some physical procedures used in ceramic research. 

/. Appl. Phys. 14, 443-450 (1943). 
Bragg, L. Seeing ever-smaller worlds. Nature 151, 545-547 (1943). 
Burton, E. F. Electron microscope studies of the capture of airborne particles by 

single fibers. Nature 152, 540 (1943). 


Fullam, E. F. Magnification calibration of the electron microscope. /. Appl. Phys. 14, 

677-683 (1943). 
Gard, S. Purification of poliomyelitis viruses. Experiments on murine and human 

strains. Acta Med. Scandinavica, Supplementum 143 (1943). 
Gard, S. Electron microscope observations on purified poliomyelitis virus prepara- 
tions. II. A contribution to the question of the epidemiology of infantile par- 
alysis.. Klin. Wochenschr. 22, 315-318 (1943). 
Gard, S. Electron microscope observations on purified poliomyelitis virus prepara- 
tions. III. A comparison with the results of physico-chemical experiments. 

Arch. Ges. Virusforschung (1943). An abstract of this article was published in 

Arkiv for Kemi, Stockholm, 17/B. I. (1943). 
Glaser, R. W. and Stanley, W. M. Biochemical studies on the virus and the inclusion 

bodies of silkworm jaundice. /. Exp. Med. 77, 451-466 (1943). 
Golz, E. Electron microscopic fine structure of fractured glass surfaces. Zeits. f. 

Physik 120, 773-777 (1943). 
Golz, E. Surface electron microscopy of glass and ceramics by the replica method. 

Zeits. f. techn. Physik 24, 8-11 (1943). 
Green, H. and Fullam, E. F. Some applications of the high resolving power of the 

electron microscope. /. Appl. Phys. 14, 332-340 (1943). 
Harker, D. and Schaefer, V. J. Metallography by electron microscope. Metal Prog. 

43, 68-69 (1943). 
Heidenreich, R. D. Interpretation of electron micrographs of silicon surface replicas. 

/. Appl. Phys. 14, 312-320 (1943). 
Henle, W. and Henle, G. Interference of inactive virus with the propagation of virus 

of influenza. Science 98, 87-89 (1943). 
Hock, C. W. and McMurdie, H. F. Structure of the wool fiber as revealed by the 

electron microscope. /. Res. Nat. Bur. Stand., Wash., 31, 229-236 (1943). 
Johnson, F. H., Zworykin, N., and Warren, G. A study of luminous bacterial cells 

and cytolysates with the electron microscope. /. Bact. 46, 167-185 (1943). 
Jungeblut, C. W. and Bourdillon, J. Electronmicrography of murine poliomyelitis 

virus preparations. /. Am. Med. Ass. 123, 399-402 (1943). 
Kinder, E. Magnesium oxide crystals in the electron microscope. Naturwiss. 31, 149 

Knaysi, G. and Mudd, S. The internal structure of certain bacteria as revealed by 

the electron microscope. A contribution to the study of the bacterial nucleus. 

/. Bact. 45, 349-359 (1943). 
Linke, F. Condensation nuclei made visible in the electron microscope. Naturwiss. 

31,230-231 (1943). 
Luria, S. E., Delbruck, M. and Anderson, T. F. Electron microscope studies of bac- 
terial viruses. /. Bact. 46, 57-77 (1943). 
Marx, T., Klemm, W., and Smekal, A. Electron microscope studies of scratches. 

Naturwiss. 31, 143-144 (1943). 
Morton, H. E. and Anderson, T. F. The morphology of Leptospira icterohemorragiae 

and L. canicola as revealed by the electron microscope. /. Bact. 45, 143-146 

Mudd, S. Changes in the bacterial cell brought about by the action of germicides 

and antibacterial substances as demonstrated by the electron microscope. Am. J. 

Public Health 33, 167-168 (1943). 
Mudd, S., Heinmets, F., and Anderson, T. F. The pneumococcal capsular swelling 

reaction studied with the aid of the electron microscope. /. Exp. Med. 78, 

327-332 (1943). 
Mudd, S., Polevitzky, K., and Anderson, T. F. Bacterial morphology as shown by the 

electron microscope. V. Treponema pallidum, T. macrodentium, and T. micro- 

dentium. J. Bact. 46, 15-24 (1943). 
Mudd, S., Heinmets, F., and Anderson, T. F. Bacterial morphology as shown by the 

electron microscope. VI. Capsule, cell wall, and inner protoplasm of pneu- 

mococcus, type III. /. Bact. 46, 205-211 (1943). 


O'Brien H C. and McKinley, G. M. New microtome and sectioning method for 

electron microscopy. Science 98, 455^456 (1943). 
Picard R G. and Duffendack, O. S. Studies on the structure of thin meta lie films by 

mean's o the electron microscope. /. AppL Pbys. 14, 291-305 (1943). 
Preckshot G. W., DeLisle, N. G., Cottrell, C E., and Katz D. L Asphaltic substances 

in crude oils. Am. Inst, of Mining and Metal, Engrs. Petroleum Division Section, 

Transactions 151, 188-205 (1943). . 

Ouarrell A. G. Electron microscope and its application to engineering problems. 

Engineer 176, 499-502; 526-528 (1943). 
Schmidt, R. W. Oxidation of metal powders in electron microscopic investigations. 

Kolloid Zeits. 102, 15-17 (1943). . , . t i . i 

Seeman, H. Migratory schlieren in electron optical pictures i of single crystals, 

especially molybdenum oxide. Natiirwtss. 31, 415-416 (1943). 
Seymour, F.J. Electron microscope, fertility, and gynecology. M. Woman s J. 

Sham'D^GiTaylor, A. R., Beard, D., and Beard, J. W. Morphology of the Eastern 
and Western strains of the virus of equine encephalomyelitis. Arch. Path. 36, 

167-176 (1943). . , . __ t _ ->, 79A 

Sjostrand, F. Electron-microscopic examination of tissues. Nature 151, 7^-/zo 

Sliepcevich, C. M., Gildart, L., and Katz, D. M. Crystals from Portland cement 
hydration; an electron microscope study. Indus*, and Engr. Chem., Ind. ed., 

Taylor, KIL^.' C, McLean, I. W., Jr., Beard D Beard, J W., Dingle, 
J. H., and Feller, A. E. Purification and character of the swine influenza virus. 
Science 98, 587-589 (1943). 

Taylor A. R., Sharp, D. G., Beard, D., Beard, J. W., Dingle, J. H., and Feller A. E. 
Iso'larion and characterization of influenza virus A (PR 8 Strain). /. Immun. 

WeisT L^J^mectrofmicrographs of bacteria medicated with penicillin. Proc. Indiana 

Acad. Sci. 52, 27-29 (1943). ' 

Weiss, L. J. Electron micrographs of rickettsiae of typhus fever. /. Immun. 47, 

Wile! 53 w! 5 J. ( and 3) Kearney, E. B. The morphology of Treponema pallidum in the 
electron microscope; demonstration of flagella. /. Am. Med. Ass. 122, 167-168 

Zworykin?V. K. Applications of the electron microscope in metallurgy Trans. Am. 

Inst Mining Met. Engrs., Inst. Metals Dw., Tech. Pub. No. 1594, 1-25 (1943). 
Fuller M. L., Bmbaker, D. G., and Berger, R. W. Mounting of pigments for electron 

microscopy. /. AppL Phys. 15, 201 (1944). 
Martin, D., Baylor, M^R. Z and Clark, G. L. A preliminary electron ^microscope 

study of the active deposit from radiothonum. Science '99, 185 (1944). 
Melnick, J. L. Detection with the electron microscope of rod-shaped P^cks in 

stools or normal and poliomyelitic individuals. J Immun. 48 25-28 (1944) 
Sharp, D. G., Taylor, A. R., McLean, I. W., Jr., Beard D Beard, J. W Feller A E 

and Dingle, J. H. Isolation and characterization of influenza virus B (Lee Strain). 

/. Immun. 48, 129-153 (1944). 


Abrams, I. M. and McBain, J. W. A Closed Cell for Electron Microscopy. /. Applied 

Physics 15, 607-609 (1944). 
Barrett, C. S. The Electron Microscope in Metallurgical Research. /. Applied 

Physics 15, 691-696 (1944). 
Benson, I. G. Electron Microscopy. General Electric Review 47, 6 (Dec, 1944). 
Calbick, C. J. Historical Background of Electron Optics. /. Applied Physics 15, 

685-690 (1944). 
Claude, A. and Fullam, E. F. An Electron Microscope Study of Isolated Mitochon- 
dria. /. Exp. Med. 81, 51-62 (1945). 
Crane, H. R. Direct Centrifugation on Electron Microscope Specimen Films. Rev. 

Sci. Instruments 15, 253 (Oct., 1944). 
Goddard, L. S. The Computation of Electron Trajectories in Axially Symmetric 

Fields. Proc. Phys. Soc. 56, 372 (1944). 
Goddard, L. S. and Klemperer, O. Electron Ray Tracing through Magnetic Lenses. 

Proc. Phys. Soc. 56, 378 (1944); 57, 63 (1945). 
Hall, C. E., Hauser, E. A., LeBeau, D. S., Schmitt, F. O., and Talalay, P. Natural and 

Synthetic Rubber Fibres. Ind. Eng. Chem. 36, 634 (July, 1944). 
Hambraeus, G. and Ranby, B. Electron Microscope Investigations of Precipitates of 

Cellulose Nitrates. Nature 155, 200 (1945). 
Heidenreich, R. D. Techniques in Applied Electron Microscopy. /. Opt. Soc. Am. 

35, 139 (1945). 
Heidenreich, R. D. and Sturkey, L. Crystal Interference Phenomena in Electron 

Microscope Images. /. Applied Physics 16, 97-105 (Feb., 1945). 
Hillier, J. and Baker, R. F. Microanalysis by Means of Electrons. /. Applied Physics 

15, 663-677 (1944). 
Katz, D. L. and Beu, K. E. Nature of Asphaltic Substances. Ind. Eng. Chem. 37, 

195-200 (1945). 
Keller, F. and Geisler, A. H. Extending Microscopic Examination of Metals. /. Ap- 
plied Physics 15, 696-704 (1944). 
Ladd, W. A. Electron Microscope Studies of Colloidal Carbon in Vulcanized Rubber. 

Ind. Eng. Chem. (Anal. Ed.) 16, 642-644 (1944). 
Levy, G. B. Dark Field Illumination in Electron Microscopy. /. Applied Physics 15, 

623-625 (1944). 
Marton, L. Stereoscopy with the Electron Microscope. /. Applied Physics 15, 726- 

727 (1944). 
Marton, L. and Hutter, R. G. E. Optical Constants of a Magnetic Type Electron 

Microscope. Proc. I. R. E. 32, 546-552 (1944). 
Marton, L. A 100-kv Electron Microscope. /. Applied Physics 16, 131 (1945). 
Picard, R. C. Studies with the Electron Microscope Diffraction Adapter. /. Applied 

Physics 15, 678^684 (Oct., 1944) . 
Porter, K. R., Claude, A., and Fullam, E. F. A Study of Tissue Culture Cells by 

Electron Microscopy. /. Exp. Med. 81, 233-246 (1945). 
Powers, D. H. Effect of Synthetic Resins on Cellulose and Protein Fibers. Ind. Eng. 

Chem. 37, 188-193 (1945). 
Williams, R. C. and Wyckofl, R. W. G. The Thickness of Electron Microscopic 

Objects. /. Applied Physics 15, 712-716 (1944). 


Name Index 

For authors cited in Bibliography, see pages 297-316. 

American Cyanamid Co., 256, 257 
Anderson, T. F, 271, 273, 274, 281 
Aristotle, 45, 53 

Bachman, C. H., 202 

Banca, M. C, 218 

Barnes, R. Bowling, 236 

Bear, R. S., 268, 270 

Behne, A., 202 

Bell Telephone Labs, 122, 182 

Bender, J. F., 218 

Bragg, Sir W., 118 

Braun, F., 87, 96 

Briiche, E., 194, 196, 200, 202, 213 

Burton, C. J., 236 

Burton, E. F., 80 

Busch, H., 165, 168, 169, 175, 178 

Calbick, C. J., 178, 182 
Caldwell, 256 
Clark, G. L., 297 
Cleomedes, 15 
Clinton, 165 

Columbian Carbon Co., 253 
Columbia Univ., 283 
Compton, A. H., 121 
Conger, P. S., 41, 253 
Crookes, Sir W., 87, 89, 91, 106 

Davisson, C. J., 122, 176, 178, 182 
De Broglie, L., 121, 122, 123, 178, 199 
Dow Chemical Co., 296 
Duffendack, O. S., 235, 297 

Eastman Kodak Co., 218, 262 
Edison, T., 99 
Einstein, A., 116 

Faraday, M., 83, 84, 94, 140 

Fermat, 177 

Flegler, 165 

Fleming, Sir A., 165 

Franklin, B., 81 

Fresnel, 84, 122 

Fullam, E., 236 

Gabor, D., 169 

Galileo, 45 

General Electric Co., 202, 205 

Germer, L. H, 122, 176, 178 
Goldstein, E., 87, 89, 90 

Hall, C. E., 196, 213, 218, 263 

Hallwachs, W., 105, 109 

Hamilton, Sir W. R., 176, 186 

Hamly, D. H., 39, 77 

Harker, D., 295 

Heidenreich, R. D., 296 

Henry, 83 

Hertz, H., 85, 109 

Hillier, J., 216, 217, 218, 271, 293 

Hittorf, J. W., 87 

Hooke, R., 63 

Houtermans, 214 

Humbert, 256 

Huyghens, C, 46, 48 


ohannson, H., 194, 195, 196, 198, 199, 

Kerr, 84 

Knaysi, 279 

Knecht, W., 200, 202, 213 

Knoll, M., 169, 171, 211, 214 

Ladd, W., 80 

Langmuir, I., 295 

Leland Stanford Univ., 216, 218 

Lenard, 91 

Levy, 245 

Lippershey, H., 45 

Lorenz, L., 87 

Luria, 283 

MacGregor-Morris, 165 

Mahl, H., 202, 213 

Marshall, 256 

Martin, L. C, 215 

Marton, L., 215, 218, 271 

Massachusetts Inst, of Technology, 268 

Matheson, L. A., 297 

Maupertuis, 177 

Maxwell, J. C, 84 

Mudd, S,, 271, 274, 275, 277, 279 

Newman, L. T., 41, 221 

Newton, I., 46, 48, 53, 106, 116, 117 



Oersted, 83, 143, 146 

Ohio Agricultural Expt. Station, 256 

Ohio State Univ., 218, 254, 255, 256, 257 

Parnum, D. H., 215 

Peck, V. G., 296 

Picard, R. H., 235 

Planck, M., 115, 116 

Plato, 45 

Prebus, A. F., 80, 216, 218, 223, 254, 255, 

256, 257 
Pythagoras, 44 

RCA Laboratories, 217, 218, 271, 281 

Radio Corp. of America, 283 

Ramberg, E. G., 295 

Ramo, S., 202 

Rankin, 163 

Rayleigh, Lord, 71, 72 

Rochow, 236 

Rockefeller Inst, of Medical Research, 

271, 279, 281 
Roemer, O., 45 
Roentgen, W., 106 
Rogowski, 165 
Ross, C. S., 256 
Riidenberg, R., 211 
Ruska, E., 169, 171, 172, 211, 215, 216 

Schaefer, V. J., 295 
Schulze, 214 


Scott, 210 

Shaw, 256 

Shoen, A. L., 263 

Siedentopf, 75 

Smith, P., 217 

Snell, W., 45 

Stanley, W. M., 271, 273, 274, 281 

Thomson, Sir J. J., 87, 90, 91, 94, 123, 147 
Thomson, G. P., 122, 178 

University of Michigan, 229 
University of Missouri, 256 
University of Pennsylvania, 271 
University of Toronto, 39, 77, 80, 216, 218, 
221, 253, 256 

Vance, A. W., 227 

von Ardenne, M., 218, 244, 245 

von Borries, B., 216 

Watson, 221 
Wehnelt, 99, 103 
Whelpton, R. V., 215 
Wiechert, E., 156, 163 

Zeiss Co., 75 

Zsigmondy, 75 

Zworykin, V. K., 201, 217, 218, 295 

Subject Index 


chromatic, 32, 245 

spherical, 32, 200, 231, 245 
Acetylene, 267 

as refracting medium, 14, 15 

in vacuum tube, 90, 96 

light rays in, 35 

sound waves in, 59, 60, 63 
Air lock, 226 
Aluminum, 63, 101 

oxide, 157, 173, 262 
Amidol, 266 
Amyl acetate, 101 
Anode, 88, 89, 103, 110, 136, 137, 195, 202, 

213, 214, 221 
Antibodies, 281 
Anti-point, 36 
Antiserum, 283 

angular, 231, 244, 245 

in electrostatic lens, 195-199, 200, 201, 
204, 214 

numerical, 39, 72, 73 
Aperture stop, 36, 166, 168, 214 
Apiezon W wax, 226 
Argon, 97 
Armco iron, 223 
Artifacts, 239, 247, 248 
Asbestos, 51, 257, 258 
Attapulgite, 58, 258 

Bacillus anthracis, 274 
Bacillus Novy, 284, 285 

and antibodies, 281 

and chemical reactions, 287 

staining of, 289, 290 
Bacteriophage, 281, 283, 285 
Barium, 101, 102, 104 

carbonate, 101 

oxide, 99, 104 
Beidellite, 258 
Bentonite, 52, 258 

magnesium, 258, 259 
Brass, 172 

Braun tube, 88, 90-93, 95, 139, 150, 156, 159, 
162, 164, 165, 214 

electrons in, 96 
Bromine, 289 

Brownian motion, 79 
Bushy stunt, 283 

Calcium, 101, 104 

carbonate, 101 

oxide, 99, 104 

pin-hole, 19, 20 

simple, 20, 21 
Carbon black, 107, 253 
Carbon dioxide, 96, 101 

monoxide, 96 

bright emitter, 201, 213 

cold, 103, 105, 159, 214 

curved, 137 

heating of, 99-102 

oxide-coated, 100-103, 150, 180, 194, 196, 
199, 201, 203, 214 

photoelectric, 110, 115, 201, 202 

rays, 87, 88, 89, 80 

shielding of, 221, 225 

tungsten, 104, 201, 203, 213 
Cathode-ray tube, 

see Braun tube, Crookes tube 
Cat's whisker, 37 
Cedar oil, 75 
Cell wall, 275, 277, 281 
Celvascene, 225, 226 
Cesium, 110 

oxide, 110 
Chlorine, 94 
Clays, 253-259 
Collagen, 268, 269 

Collodion, 239, 240, 245, 262, 273, 283, 295 
Compass, 139 

Condenser lens, 222, 224, 229, 235 
Copper, 63 
Corpuscular theory, 34, 45, 48, 53, 116, 

117, 121, 122 
Corynebacterium diphtheriae, 278 
Cosmic rays, 85, 97 
Crookes tube, 87, 135, 165 
Crystallization, 256, 258, 262, 263, 265, 289 
Cuprene, 267 
Cyclotron, 226 

Dark-field technique, 244 
Dehydration of specimen, 244, 247, 251 





and light, 64 

and refraction, 15, 16 

and sound, 61, 63 

of specimens, 246, 289 
Diamond, 89 

Diatom, 24, 33, 39, 41, 77, 234, 251, 253 
Dickite, 258 

Diffraction, 34, 35, 53, 54, 122, 177 
Dyes, 42 

Ebonite, 81, 95, 129 

Elastic constant, 61, 62, 63, 64 

Elasticity, 61 

Electric field, see Field, electric 

Electricity, 81, 82, 83, 84, 94, 130, 131 

Electrolysis, 94, 102 

Electromagnet, 145, 146 

Electromagnetic theory, 81-85 

see also Wave theory 
Electron analyzer, 249 
Electron gun/ 204, 221, 235, 244 
Electron optics, 165, 166, 176, 188, 192-195 
Electron microscope, 
electrostatic, 194-205, 210 
resolving power of, 199, 201, 202 
General Electric, 202-205 
emission type, 207, 210, 211 
magnetic, '207, 211-218, 221-227 
disadvantages of, 251 
first, 119 

focus depth of, 251 
resolving power of, 251 
universal model (RCA), 249 
transmission type, 210 
vs. optical, 208, 209 
Electron volt, 134 

and light, 105, 106, 109, 110, 111, 115 

121, 122 
and matter, 95, 96, 177 
charge of, 93, 94 
definition of, 90, 148 
focussing of, 124, 127, 155, 156, 159-17f 

from cathodes, 100, 101, 102, 103, 105 
from metals, 97, 99, 100 
in force fields, 134-137, 143-150, 154, 

mass of, 92, 93, 94, 95 
mean free path of, 226 
primary, 105 

scattering of, 224, 246, 285 
secondary, 104, 105 
velocity of, 135, 137 
wave length of, 123 
Electrostatic lens, 124, 178-188 
see also Lens, electrostatic 

Energy, see also Light 

and distance, 113, 114 

and frequency, 116 

and sound, 60 

conservation of, 115 

electrical, 82, 84 

in cathode tubes, 97, 103, 104 

in electric fields, 133 

kinetic 103, 104, 106, 133 

nature of, 116 

radiant, see Light 

transmission of, 48, 49, 53, 55, 92 
Enlargement, photographic, 248 
Equipotential lines, 132, 134, 136, 137 
Ether, 45, 64, 69, 70, 71, 83, 84, 87 
Ethylene dichloride, 240 

and lenses, 19, 29, 30, 32 

and camera, 21 

and glasses, 25 

limitations of, 22, 23 

resolving power of, 233 

Face powder, 37 
Field, see also Force 

electric, 83, 84, 90, 92, 93, 99, 124, 132, 
133, 148, 176 
focussing by, 178-188, 205 

magnetic, 83, 84, 90, 92, 124, 139, 142- 
150, 154 
alternating, 151 

focussing by, 155, 159-172, 222-224 
Filament, tungsten, 203, 216, 224, 225 
Films, mounting specimens on, 240, 262 
Flagella, 275, 281 
Fluorescent screen, 88, 90, 91, 97, 159, 166, 

192, 195, 205, 210, 214, 222, 235, 244 
Fluorspar, 89 
Focal length, 27 
Focus, 17, 18, 20, 21, 22, 26, 27, 29 

depth of, 246, 251 

of cathode rays, 90, 91 

by electrostatic field, 178-188, 205 

by magnetic field, 124, 137, 155, 156, 
159-172, 222-224 
Force, see also Field 

definition of, 93, 127, 128 

gravitational, 129, 139 

electrical, 82, 83, 93, 99, 130, 132, 139 

lines of, 131, 132, 134, 135, 142 

magnetic, 82, 83, 139, 142, 143-150 

vectors, 151-153, 187 
Formvar, 239, 240, 248, 262, 296 

definition of, 56 

of sound, 61, 122 

of light, 111, 112, 113, 116, 117 

threshold, 112 




in discharge tubes, 89, 91, 96, 97 

wave motion in, 59, 60, 70 
Geissler tube, 87, 88 
Geometrical optics, 175 

as insulator, 95, 226 

as refracting medium, 16, 17, 18, 186 

charged, 81, 129 

fluorescence of, 89 

in lenses, 17, 26, 35, 161, 162, 165, 168, 

in lens mounting, 195 

magnifying, 29 

ruby, 78 
Glycerin, 186 
Gold, colloidal, 78 
Grease, 24 
Grid, 195 
Grundite, 258 

Halloysite, 255, 258 
Heat, 44 

and electron emission, 99, 100, 102 
Hydrogen, 63, 64, 91, 101 

atom of, 94, 95, 96 
Hydroquinone, 266 
Hypervac pump, 227 
Hyvac pump, 226 

Iceland spar, 46, 84 
Hike, 258, 259 

and diffraction, 35 

and lens, 17, 26, 27, 28 

blurring of, 247 

formation of, 20-23, 191, 192, 196, 202, 
210, 214, 224 

distortion of, 248 

in camera, 19, 20, 21 

in eye, 22, 23, 25, 26, 30 

inverted, 28, 29 

in microscope, 30, 31 
Immersion-objective lens, 194, 195, 200, 201 
Influenza virus, 86 
Interference, 57, 67-71, 122 
Invar, 236 
Ionization, 96 

Iron, lens shielded by, 168, 169 
Iron oxide, 140 

Latex, 118, 252 
Lead, 63 

aberrations of, 32 
and prisms, 18 
camera, 20, 21 

condenser, 222, 224, 227, 235 
converging, 26, 28, 29 
diverging, 27, 28 
electrostatic, 124, 178-188 
aberration of, 200 
3-electrode, 196 
4-electrode, 198 
focussing of, 205 

immersion objective, 194, 195, 200-202 
negative, 192, 193 
positive, 194 

resolving power of, 199, 201, 202 
eye, 22, 23, 25, 26, 32 
focal length of, 27 

glass, 17, 26, 35, 161, 162, 165, 168, 207 
magnetic, 124, 159-172, 223, 224, 227, 235 
aberrations of, 245 
resolving power of, 215, 216, 251 
power of, 19 
objective, 30, 33, 227, 235 
projector, 30', 227, 235 
resolving power of, 71 
reading, 29 

and electricity, 81-85 

and electrons, 89, 90, 105, 106, 109, 110, 

111, 115 
and lenses, 25-29 
corpuscle, 116, 117, 121, 122, 176 
diffraction of, 34, 35, 53, 54, 122, 177 
in camera, 20, 21 
intensity of, 113, 114 
interference of, 57, 67-71, 122 
nature of, 34, 35, 177 
polarization of, 46, 47, 71, 84 
refraction of, 13, 14, 15, 44, 183, 184 
scattering of, 75, 78 
speed of, 46, 64 

theories of, 44-49, 70, 84, 115, 118 
ultraviolet, 75 
wave lengths of, 85 
Lippman crystals, 263, 265, 266 
Lodestone, 139, 140, 146 

Kaolin, 52 

Kaolinite, 254, 258, 259 
Kenetron, 227 
Krypton, 97 

Magnesium, 101 
oxide, 125, 262 
Magnetic deflection, 150 
Magnetic field, see Field, magnetic 



Magnetic lens, 124, 159-172, 215, 216, 223, 
224, 227, 35 

aberrations of, 245 

resolving power of, 215, 216, 251 
Aiagnetism, 81, 82, 83, 84 
Magnets, 139-142, 145, 146, 147 

and lenses, 28, 32, 235 

and resolving power, 74, 75, 231 

by microscope, 30, 31 

electrostatic, 192, 196-199, 202, 205 

magnetic, 210, 215, 216 

useful, 231, 235 
McLeod gauge, 226 
Mechanical models, 43 
Methylene blue, 98 
Metal, 266 

interpretation of, 246, 247, 251 

production of, 241, 242 

stereoscopic, 232, 240, 245, 246, 291, 293 

compound, 30, 31 

electron, see Electron microscope 

resolving power of, 33, 34, 71, 72 

simple, 29 
Microtome, 273, 290 
Midge, 12, 239, 253 
Mine dust, 259, 260, 261 
Molecules, organic, 266, 283 
Molybdenum, 102, 104, 195 

as cathode, 201 
Montmorillonite, 58, 258 
Morphology, 273, 279 

Neon, 97 

Nickel, 101, 102, 104 

Nitrogen, 96 

Nitroglycerin, 101 

Nitrous oxide, 101 

Mycobacterium tuberculosis, 277 

Nontronite, 258 

Numerical aperture, 39, 72, 73 

Objective lens, 30, 33, 227, 235 
Osmic acid, 270 
Oxygen, 96, 101, 102, 110 

Paraphenylenediamine, 266 

Periplast, 281 

Planets, 79 

Platinum, 102, 104 

Phosphotungstic acid, 268, 269 

Photo cell, 109, 110, 111 

Photoelectric cathode, 110, 115, 116, 201, 

Photoelectric effect, 106, 109, 112, 116, 118, 


Photographic plate, 20, 30, 91, 106, 165, 

166, 210, 214, 224, 226, 231, 295 
Photon, 117, 121, 122 
Physical optics, 175 
Polarization, 46, 47, 71, 84 
Polaroid, 47, 71 
Pollen dust, 156, 251 
Polystyrene, 241, 242, 243, 293, 296 
Potential difference, 130, 132, 134, 136 
Potential gradient, 187 
Pneumococcus, 275, 276 
Power, 127, 128, 129 

water, 135 
Prism, 16, 17, 18, 46 
Projector lens, 30, 227, 235 
Proteins, 268 
Protons, 95, 96 
Protoplasm, 275, 277, 281, 290 
Pulp, paper, 42 
Pumps, in electron microscope, 226, 227 

Quantum theory, 115-118 

Radiation, see Light 

Radon, 97 

Rays, light, see Light 

cathode, see Cathode rays 
Refraction, 13, 14, 15, 16, 44, 53, 184, 185 
Refractive index, 177, 184, 185, 200 
Resolving power, 33, 34, 36 

and wave length, 71, 72 

and magnification, 74, 75, 231 

of eye, 233 

of electrostatic lens, 199, 201 

of magentic lens, 215, 216, 251 
Replicas, 189, 236, 229, 239, 243, 292, 294 

how made, 240, 241, 293 
Retina, 21, 22, 23, 25, 26, 48, 79, 166 
Rickettsia, 271, 282 
Ruby glass, 78 

Shadow technique, 237 

of cathode, 221, 225 

of lens, 168, 169 

of clays, 258, 259 

of specimens, 247 
Silica, 241, 242, 243, 259, 296 
Silicosis, 259 
Silver, 295 

in emulsions, 262, 263, 265, 266 

in photo cell, 109, 110 
Silver bromide, 91, 106, 263, 265, 266 
Silver oxide, 110 
Smallpox virus, 279, 280 
Snell's law, 184, 186 
Sodium chloride, 232, 248, 262 



Sodium thiosulfate, 263 

Solenoid, 145, 146, 150, 155, 159, 162, 166, 

168, 210, 214, 215 
Solids, wave motion in, 62, 63 
Sound, 59, 60, 61, 62 

speed of, 61, 63 

changes in, 274, 290 

deformation of, 248 

dehydration of, 244, 247, 251 

density of, 246, 289 

mounting of, 239, 241 

preparation of, 239 

shrinkage of, 247 
Spectrum, 46 

visible, 73 
Spherical aberration, 32, 231, 245 
Spirochete, 279, 281 
Spurious disk, 36, 71, 72 
Staining, 289, 290 
Steel, 63, 64 

Stereoscopic technique, 291, 293 
Stokes' law, 256 
Strontium, 99, 104 

carbonate, 101 

oxide, 104 
Sulfur, 95 

Surface area, 253, 258 
Sylphon bellows, 226, 227 

Tellurium, 289 

Thermal precipitator, 259, 260 

Thermionic emission, 100, 101, 102, 103 

Titanium, 101 

Tobacco mosaic, 272, 273, 281, 283 

Tooth, 242, 243 

Trachea, 57, 65, 251 

Treponema pallium, 281 

Tuberculosis bacillus, 275, 286 

Tungsten, 99, 102, 103 

as cathode, 201, 203, 213 

thoriated, 104 
Typhus, 282 

Ultramicroscope, 75, 78 

Vacuum, 91, 96, 97, 124, 214, 225, 226 

Vectors, 151, 152, 153, 187 

Vibrio comma, 157, 157, 290 

Vibrio schuylkiliensis, 238, 277 

Vinylchloride, 267 

Viruses, 271, 273, 274, 275, 281, 283 

Vision, 13, 44, 45 

defective, 25 

limits of, 22, 23 
Voltage, accelerating, 227 


as refracting medium, 14, 15, 184, 185 

difference of level, 129, 135 

evaporation of, 100 

in clays, 258, 259 

mounting specimens in, 239, 241 

splashing of, 104, 105, 106 

velocity of sound in, 61, 63 

waves, 49, 53, 54, 55, 56 
Wave length, 

and color, 74 

and frequency, 113, 122 

and interference, 69 

and resolution, 72, 73 

classification of, 85 

definition of, 55 

in photo cell, 111, 112 

of electrons, 123 
Wave theory, 68, 70 

limitations of, 113, 115 

vs. corpuscular theory, 117, 121, 122 

elastic, 61, 71 

in solids, 62, 63 

light, 45, 48, 54, 68, 70, 81 
and quantum theory, 115-118, 121, 122 
associated, 117, 122 
interference of, 67, 67-70, 71 
properties of, 85 
speed of, 64 

longitudinal, 60, 61, 62, 71 

sound, 59-63 

transverse, 55, 62, 71, 84 

water, 49, 53-56 
Willemite, 89, 106, 139 
Wool fiber, 189 

Work, 127, 128, 129, 130, 132, 134, 136 
Work function, 103 

Xenon, 97 

X-ray, 85, 106, 256, 268, 273 
diffraction studies, 270 

Zincblende, 89 
Zinc-cadmium sulfide, 222 
Zinc oxide, 124, 262 

SC IENCE &m£>ue 




Electron microscope main 


3 12b2 D2157 772D